SHIFTWORK...Shiftwork, including night work, involved more than 17% of the total European Union (EU)...

SHIFTWORK

–563–

SHIFTWORK

1. Definition and Occurrence of Exposure

1.1 Definition of shiftwork

The International Labour Office (International Labour Organization, 1990a) defines working in shifts as “a method of organization of working time in which workers succeed one another at the workplace so that the establishment can operate longer than the hours of work of individual workers.”

The European Council Directive 93/104 (1993) declares that “concerning certain aspects of the organisation of working time, shiftwork shall mean any method of organising work in shifts whereby workers succeed each other at the same work stations according to a certain pattern. Shiftworker shall mean any worker whose work schedule is part of shiftwork.”

Besides these definitions, in the scientific literature, the term “shiftwork” has been widely used and generally includes any arrangement of daily working hours other than the standard daylight hours (7/8 am – 5/6 pm).

In most cases, shiftwork is synonymous of irregular, odd, flexible, variable, unusual, non-standard working hours.

1.2 Types of shiftwork

Several types of shiftwork exist and can be described as follows: (a) permanent – people work regularly on one shift only, i.e. morning or

afternoon or night; or rotating – people alternate more or less periodically on different shifts;

(b) continuous – all days of the week are covered; or discontinuous – interruption on weekends or on sundays;

(c) with or without night work – the working time can be extended to all or part of the night, and the number of nights worked per week/month/year can vary

564 IARC MONOGRAPHS VOLUME 98

considerably. Moreover, the definition of “period of night work” varies from country to country, i.e. in some countries it ranges from 8, 9 or 10 pm to 5, 6 or 7 am, and in many others from 11 or 12 pm to 5 or 6 am (See Table 1.1).

Table 1.1. Definitions of night work and night worker in some European countries

COUNTRY NIGHT TIME/NIGHT WORK

NIGHT WORKER

AUSTRIA Night work: period between 22:00 and 05:00

The workers who work at least 3 hours between 22:00 and 05:00 on at least 48 nights per year (EU-Nachtarbeits-Anpassungsgesetz 2002)

BELGIUM Night work: a period, generally of 8 hours, between 20:00 and 06:00

Loi du 17/02/1997 et Loi du 04/12/1998: Act of 17 February 1997

FINLAND Night work: Work carried out between 23:00 and 06:00

Night shift refers to a work shift with at least 3 hours of duty between 23:00 and 06:00 (Working Hours Act 605/1996)

FRANCE Night time: a period between 22:00 and 05:00 Night work: whichever work period between midnight and 05:00

Any employee working usually at least 2 times per week for at least 3 hours over the period defined as night work (Loi 461/1998)

GERMANY Night time: the time between 23:00 and 06:00 (in case of bakers between 22:00 and 05:00). Night work: all work which occupies more than 2 hours of night time

“Night workers” means workers who usually work nights on rotating shifts schedules, or work at night for not less than 48 days in a calendar year (Arbeitszeitgesetz 1994)

GREECE Night time: a period of 8 hours which includes the period between 22:00 and 06:00

A worker who during night time works at least 3 hours of his/her daily working time or a worker who has to perform night work for at least 726 hours of his/her annual working time (Presidential Decree n. 88/1999)

IRELAND Night time: period between midnight and 07:00

a) an employee who normally works at least 3 hours of his/ her daily working time during night time; b) an employee whose working hours during night time, in each year, equals or exceeds 50 per cent of the total number of hours worked during the year (Statutory Instruments n. 485/1998)

SHIFTWORK 565

Table 1.1 (contd)

COUNTRY NIGHT TIME/NIGHT WORK

NIGHT WORKER

ITALY Night work: the activity carried out in a period of at least 7 consecutive hours comprising the interval between midnight and 05:00

a) any worker who during the night period carries out, as a normal course, at least 3 hours of his/her daily working time; b) any worker who during the night period, carries out part of his/her daily working time as defined by collective agreements; in default of collective agreements, any worker who works at night at least 80 working days per year (D.Lgs. 66/2003)

NETHERLANDS Night work: work which covers all or part of the period from midnight to 06:00

PORTUGAL Night time: a period between 20:00 and 07:00

a) any worker who works at least 3 hours during the night period; b) any worker who during the night period, carries out part of its daily working time as defined by collective agreements (Decreto Lei 73/1998)

SPAIN Night time: a period which includes the interval between 22:00 and 06:00

A worker who at night carries out at least 3 hours of his/her daily working time (Real Decreto Lei 1/1995)

SWEDEN Hours between midnight and 05:00

A worker that works at least 3 hours of his/her daily work during night time, or a worker that most likely will work at least 38% of his/her annual work during the night (Working Hours Act 1982)

UK Night time: a period lasting not less than 7 hours, and which includes the period between midnight and 05:00

A worker who, as a normal course, works at least 3 hours of his/her daily working time during night time, or who is likely, during night time, to work at least such proportion of his annual working time as may be specified for the purposes of these Regulations in a collective agreement or a workforce agreement (Statutory Instrument No.1833/1998).

Table compiled by the Working Group

The shift systems can also differ widely in relation to other organizational factors: (a) length of shift cycle – a “cycle” includes all shifts and rest days lasting as long

as the series of shifts restart from the same point; there can be short (6–9 days), intermediate (20–30 days), or long (up to 6 months or more) cycles.


(b) duration of shifts – in general, the length of a shift is 8 hours, but can range from 6 to 12 hours.

(c) number of workers/crews who alternate during the working day. (d) start and finish time of the duty periods. (e) speed of shift rotation – this depends on the number of consecutive days

worked before changing shift. It can be fast (i.e. every 1, 2 or 3 days), intermediate (i.e. every week), or slow (i.e. every 15, 20 or 30 days). This factor has considerable influence on the number of consecutive night shifts and rest days.

(f) direction of shift rotation – it can be clockwise (i.e. morning/afternoon/night) or counter-clockwise (i.e. afternoon/morning/night) with consequent different duration of the intervals between shifts. Clockwise rotation is also referred to as “phase delay” or “forward rotation,” and counter-clockwise rotation, “phase advance” or “backward rotation”. They have a different impact on the adjustment of the circadian rhythm.

(g) number and position of rest days between shifts. (h) regularity/irregularity of the shift schedules. All of these factors can be combined in different ways depending on the demands

specific to the occupation. In the industrial sectors (i.e. mechanical and chemical), shiftwork is usually arranged

in continuous three-shift systems. A similar number of crews/workers work both on day and night shifts, with regular shift schedules either on fast or slow rotating cycles, with fixed start and finishing times.

In the transport sector, schedules are often quite irregular, both in terms of number of consecutive shifts, shift rotation, start and finishing times, duration of the duty periods, location, and amount of rest days.

In the health-care sector, quite different shift schedules are operated with different rotation (clockwise or counter-clockwise), variable start and finishing time, and different amount of night shifts.

In the service sector, workers are commonly employed on split shifts, for example, very early morning and late afternoon shifts in road- and office-cleaning, merchandise delivery, or permanent night work (security guards).

In the leisure sector, work is mainly performed during the late afternoon and night hours, with a long duration of shifts.

Different shiftwork systems have potentially different impacts on the health of the workforce, disturbing the circadian rhythm, an essential biological function, in different ways, and also inducing sleep deprivation (see Section 4). In addition to shiftwork schedules, other factors can affect tolerance to shiftwork and night work such as individual characteristics, family situation, social conditions, and working conditions (Fig. 1.1; Costa et al., 1989; Costa, 1996, 2003; Knauth, 1996; Knauth & Hornberger, 2003).

SHIFTWORK 567

Working conditions Compensative measures Monetary compensation

Work organization Job satisfaction

Work load Counselling

Social conditions Shiftwork tradition Community organization Social involvement Social support Commuting Public services

Working hours Shift schedules

Timetables Overtime

Amount of night work Flexible times arrangement

Figure 1.1. Factors that can affect tolerance to shiftwork and night work (Costa, 2003)

1.3 Occurrence of shiftwork

Increasingly, shiftwork and night work are becoming more common in our so-called “24-hour” (or “24/7”) society. Shiftwork and night work enable round-the-clock activities required for meeting technological needs (e.g. power plants, oil refinery, and steel industry), social services/utilities functions (e.g. hospitals, transports, police and security forces, firefighting, hotels, and telecommunications), productive and economic demands (e.g. textile, paper, food, mechanical, and chemical industry), and the needs of the leisure industry.

More than two and a half billion people are officially recognized as workers according to the most recent statistics of the International Labour Organization (International Labour Organization, 2006), two-thirds of which in the Asiatic continent. Reliable data on the numbers of workers employed in shiftwork is not easy to collect due to the lack of robust statistics in many countries, and/or differences in methods of data collection not always being comparable.

Family and living conditions Marital status Number and age of children Partner’s (shift)work Housing conditions Family attitudes Incomes

Individual characteristics Age

Gender Circadian structure

Personality and behaviours Sleep strategies State of health


However, in Europe, the European Foundation for the Improvement of Living and Working Conditions has been carrying out periodical surveys on working conditions, including working hours, every 5 years since 1990. According to the third survey, carried out in 2000 in 15 European countries and involving 21703 workers, people who do normal or standard daytime work (that is those who do not work more than 40 hours per week, more than 10 hours per day, on shifts, at night, on sundays and/or saturdays, and part-time) represented only 24% of the whole population, 27% of employed workers, 8% of self-employed workers, with men and women sharing the same proportion (24%) (Costa et al., 2004).

According to the results of the fourth survey carried out in 2005 (European Foundation, 2007), the weekly working hours among the 31 European countries examined ranged from an average of 34 hours in the Netherlands to 55 hours in Turkey, and from a minimum of 8 hours (as part-time work) to a maximum of 90 hours (as overtime work). Shiftwork, including night work, involved more than 17% of the total European Union (EU) working population (Table 1.2), with large variations among countries, and between old and new member States (from 6.4% to 30%). There were also quite large differences among EU countries when looking at evening (from 36% to 58%) and night work (from 18% to 24%) (Fig. 1.2). Evening and night work are mostly used in the hotel and restaurant industry, health care, and transport and communication sectors, usually employing an older workforce (Fig. 1.3). More generally, shiftwork in its different definitions is used by one-third of people working in the health-care sector and the hotel and restaurant industry, and in one fourth of cases in the manufacturing, transport, and communication sectors (Table 1.3). According to age and gender (Table 1.4), the average percentage of shiftwork including night work is quite similar in both men and women, with quite a high percentage of workers aged over 55 employed in night work (10.5%).

In the USA, according to the Bureau of Labor Statistics (US Bureau of Labor Statistics, 2005), in 2004, almost 15% of full-time salaried workers usually worked on alternate shifts. Men were more likely than women to work such shifts (16.7% and 12.4%, respectively). This was also true for the black population when compared to the caucasian, hispanic or latino, or asian populations, with shiftwork progressively decreasing with increasing age (Table 1.5). The prevalence of shiftwork was greatest among workers in the service industry (32.6%; Table 1.6), particularly the protective service industry (50.4%, includes police, firefighters and guards), food preparation and serving (49.4%), and those employed in production, transportation, and material-moving occupations (29%). The proportion of workers on alternate shifts was highest in the leisure and hospitality (45.8%), mining (31.5%), and transportation and utilities (27.8%) industries.

SHIFTWORK 569

Table 1.2. Prevalence (%) of shiftwork that includes night work, by country in Europe in 2005 (4th EU Survey on working conditions)

Austria 13.2 Belgium 13.2 Bulgaria 21.0 Croatia* 33.5 Cyprus 11.8 Czech Republic 22.2 Denmark 9.3 Estonia 20.4 Finland 24.3 France 14.9 Germany 15.7 Greece 13.0 Hungary 20.7 Ireland 12.0 Italy 18.1 Latvia 21.9 Lithuania 19.4 Luxembourg 13.9 Malta 22.3 Netherlands 11.8 Norway 23.4 Poland 10.3 Romania 21.0 Slovakia 27.5 Slovenia 30.0 Spain 22.2 Sweden 16.0 Switzerland 12.9 Turkey* 6.4 United Kingdom 15.4 EU27 17.3 EU25 17.1 EU15 16.0 NMS 23.0

EU27: 25 EU Member States, plus the two countries that joined the European Union in 2007 – Bulgaria and Romania EU25: 15 EU Member States. plus the 10 new Member States that joined in 2004 EU15: 15 EU Member States prior to enlargement in 2004 NMS: 10 New Member States that joined in 2004 * Two candidate countries for membership of the EU: Croatia and Turkey


Figure 1.2. Prevalence of evening and night work by group of country in Europe in 2005 (4th EU Survey on working conditions)

Country codes EU15 15 EU Member States prior to enlargement in 2004 NMS 10 new Member States that joined in 2004 EU25 15 EU Member States, plus the 10 NMS EU27 25 EU Member States, plus the AC2 AC2 Two countries that joined the European Union in 2007: Bulgaria and Romania CC2 Two candidate countries for membership of the EU: Croatia and Turkey AT Austria BE Belgium BG Bulgaria CY Cyprus CZ Czech Republic DK Denmark EE Estonia FI Finland FR France DE Germany EL Greece HU Hungary IE Ireland IT Italy LV Latvia LT Lithuania

LU Luxembourg MT Malta NL Netherlands PL Poland PT Portugal RO Romania SK Slovakia SI Slovenia ES Spain SE Sweden UK United Kingdom HR Croatia NO Norway CH Switzerland TR Turkey

SHIFTWORK 571

Country groups Continental countries: AT, BE, DE, FR, LU Ireland and the United Kingdom: IE, UK Eastern European countries,: CZ, EE, HU, LT, LV, Pl, SI, SK Southern European countries: CY, EL, ES, IT, MT, PT Scandinavian countries and the Netherlands: DK, FI, NL, SE Acceding countries: BG, RO Candidate countries: HR, TR EFTA (European Free Trade Association): CH, NO Typology adapted from Esping-Andersen

Figure 1.3. Prevalence of evening and night work by work sector in Europe in 2005 (4th EU Survey on working conditions)


Table 1.3. Prevalence (%) of shiftwork that includes night work, by work activity in Europe in 2005 (4th EU Survey on working conditions)

Agriculture and fisheries 4.5 Armed forces 19.2 Clerks 13.4 Construction 5.3 Craft and related trades 17.6 Education 8 Electricity, gas and water supply 17.4 Elementary occupations 19.2 Financial intermediation 6.2 Health 35.5 Hotels and restaurants 29.9 Legislators, senior officials and managers 8.8 Manufacture and mining 25.8 Plant and machine operators and assemblers 34.5 Professionals 11.6 Public administration and defence 17.7 Real estate 9.5 Service, shop and market sales workers 26.9 Skilled agricultural and fishery workers 2.6 Technicians and associate professionals 14.3 Transport and communications 24.1 Wholesale and retail trade 16.3 Self-employed 5.7 Employee 19.8

Table 1.4. Prevalence (%) of shiftwork, including night work, by gender and age, in Europe in 2005 (4th EU Survey on working conditions)

Gender Men 17.2 Women 17.4 Age (years) ≤24 20.7 25–39 19.1 40–54 16.7 ≥55 10.5

Table 1.5. Percent distribution of shiftwork in full-time wage and salary workers by sex, race and ethnicity, in the USA in 2004 (US Bureau of Labor Statistics)

Total (>16 years) 14.8 Men 16.7 Women 12.4 White 13.7 Black or african american 20.8 Asian 15.7 Hispanic or latin ethnicity 16

SHIFTWORK 573

Table 1.5 (contd)

20–24 years 22.3 25–34 years 15.2 35–44 years 14.1 45–54 years 12.8 55–64 years 12.5 ≥65 years 10.3

Table 1.6 Percent distribution of shiftwork in full-time wage and salary workers, by occupation and industry, in the USA in 2004 (US Bureau of Labor Statistics)

Occupation

Management professionals 8.7 Service occupations 36.1 Sales and office occupations 16.4 Natural resources, construction and maintenance 7.6 Production, transportation and material-moving occupations 26.4

Industry Private sector 15.4 Agriculture and related industries 9.5 Mining 31.5 Construction 2.8 Manufacturing 17.7 Wholesale and retail trade 22.0 Transportation and utilities 27.8 Information 15 Financial activities 7.0 Professionals and business services 9.4 Education and health services 12.8 Leisure and hospitality 45.8 Other services 13.0 Public sector 11.9 Federal government 14.7 State government 11.5 Local government 11.3

1.3.1 Exposure assessment

It is difficult to assess the effective “exposure” and the consequent “risk” of shiftwork with the common methods used (i.e. in toxicology) as the “dose” can widely differ not only in terms of quantitative load, i.e. in relation to the time spent in shiftwork, but mainly in terms of qualitative aspects, i.e. in relation to the interference that different shift


systems may have on biological and psychosocial functions, also taking into account several concurrent individual, social, and working factors.

The various combinations of these aspects can cause a different amount of stress and also different stress-related effects, thus making it difficult to compare groups without adjusting for the amount of “exposure”, at least for the type of shift schedule and the years spent in shiftwork.

From a biological perspective, the occurrence and amount of night work is the most important factor to be considered. It is then possible to estimate roughly the effects (more or less severe) the different shift systems may have on health through interference on biological function, and on psychosocial issues.

Several methods have been proposed for assessing working time arrangements to evaluate their potential risk for health and well-being. The criteria most widely used are perturbation of the circadian rhythm, performance at work (ability to work efficiently), health, and social life (Wedderburn, 1994).

The “Rota Risk Profile Analysis,” proposed by Jansen and Kroon (1995), describes several risk factors associated with with roster design, related to both physiological and psychosocial aspects, that must be considered. in particular: regularity of shift timetable, periodicity (i.e. the degree to which the “biological clock” is disturbed), shift load (i.e. the average length of shifts) and week load (i.e. the average length of the working week), opportunities for night rest (for sleeping between 11 pm and 7 am) and constancy in night rest (variation in the week), predictability of the shift cycles, opportunities and constancy for household and family tasks, opportunities and constancy for evening recreation (between 7 pm to 11 pm), opportunities and constancy for weekend recreation.

1.3.2 Factors influencing shiftwork exposure and health

Many health impairments associated with shiftwork have been reported. These include psychosomatic disorders of the gastrointestinal tract (colitis, gastroduodenitis, and peptic ulcer) and of the cardiovascular system (hypertension, ischaemic heart diseases), as well as metabolic disturbances, that are influenced by other time- and work-related factors and behaviours (Costa, 1996; Knutsson, 2003).

About 20% of all workers have to stop shiftwork altogether after a very brief period because of serious health problems, 10% do not complain about shiftwork during their whole working life, while the remaining 70% withstand shiftwork with different levels of intolerance that can become more or less manifest at different times and with different intensity in terms of discomforts, troubles or diseases (Waterhouse et al., 1992).

1.3.3 Some lifestyle factors that possibly modify the effects of exposure

Some personal risk factors can act either as confounders or mediators, and/or modifiers, of the relation between shiftwork and health. Smoking and diet, generally considered as confounders in epidemiological studies, can also be intermediate factors of

SHIFTWORK 575

the effects of shiftwork (i.e. for cardiovascular and gastrointestinal disorders). Many studies have reported that shiftworkers tend to smoke more (Bøggild and Knutsson, 1999; van Amelsvoort et al., 2006) and/or increase their consumption of caffeinated or alcoholic drinks at night, as well as modify the composition and the caloric distribution of the different meals, i.e. by increasing carbohydrate intake at regular intervals (Reinberg et al., 1979; Romon et al., 1986; Lennernäs et al., 1993). Metabolic disturbances have been found to be prevalent in shiftworkers (Knutsson et al., 1990; Karlsson et al., 2001). Of concern are mainly the risks for cardiovascular disease and obesity (Tenkanen et al., 1998).

1.3.4 Specificity of exposure to shiftwork for some particular occupations

(a) Aircraft crew and transmeridian travel over time zones

Aircraft crews operating on long transmeridian flights have to cope with a shift in external time in addition to the shift of the working period. Therefore, the individual biological rhythms have to adjust to abnormal working hours in a changed environmental context. The short-term problems arising from these conflicts are similar to those of normal shiftwork, but are often aggravated by the fatigue due to the extended duty periods, and by a loss of the usual external time cues.

After a long transmeridian flight, the circadian system does not adjust immediately to the new local time, but requires several days in relation to the number of time zones crossed; the greater the number, the longer is the time required, considering that the human circadian system can adjust to no more than 60–90 minutes per day (Wegmann and Klein, 1985).

The adjustment is generally more rapid in westbound (about 1 day per hour of shift) than eastbound flights (about 1.5 day per hour of shift; Ariznavarreta et al., 2002; Gander et al., 1989; Suvanto et al., 1990). In the first case, there is a progressive phase delay of the circadian rhythm in relation to the extended personal day, whereas in the latter there is a phase advance due to the compressed day (directional asymmetry). A complete readjustment after transition of six time zones was found to take 13 days and 10 days in eastward and westward flights, respectively (Wegmann and Klein, 1985).

In addition, crews are exposed to many other concurrent risk factors, such as cosmic radiation, electromagnetic fields, lighting, noise, acceleration, vibration, mental stress, fixed postures, and pressurization.

No statistics are currently available on the entire population employed in transmeridian flights, and consequently in related shiftwork, which is generally characterized by very irregular shift schedules. Only in the case of pilots and flight engineers, are there data that can provide a rough idea of the possible number of workers involved, considering that they generally account for about 20% of the total aircraft crew members.


The US Aircraft Owners and Pilots Association (IAOPA 2007) estimated that the civil aviation worldwide during 2004 consisted of approximately 370 000 aircraft and 1.3 million pilots flying some 39 million hours. On balance, roughly 600 000 pilots were employed in commercial air transportation worldwide (including cargo and charter).

The US Bureau of Labor Statistics (2007), reported that civilian aircraft pilots and flight engineers held about 107 000 jobs in the USA in 2006. About 79 000 worked as regular airline pilots, copilots, and flight engineers. The remainder were commercial pilots who worked as flight instructors at local airports or for large businesses that fly company cargo, and executives in their own airplanes or helicopters.

(b) Watchkeeping and driving

Ship’s crew members engaged in long distance navigation work on continuous shiftwork, with some differences compare to land-based shiftworkers. For example, they can only take rest time in their place of work after the duty period, and usually have no rest days up until the end of the sea voyage is concluded. Moreover, they also have to cross several time zones (at different speed compared to flight crews), and their leisure time is limited both in terms of space and time. Several different shift systems are used. In merchant fleets, the personnel is generally divided into two or three crews working 12-hour or 8-hour shifts respectively, whereas on warships the crew work more frequently on the “4-hour watch” system, by dividing the 24-hour period into six 4-hour watches, and rotating on a “4-h on/8-h off” schedule, that allows one full night sleep in three. In general, in this shift schedule, the average amount of sleep is nearly the same as that of dayworkers, but the sleep is fragmented into two periods. A further system, the 6-hour on/6-hour off system is becoming more and more common on warships. However, high irregularity and variability of shift duration and rotation are quite frequent due to crew shortage, additional duties and unexpected situations, thus the amount of rest and sleep hours may vary considerably among days and subjects (Eriksen et al., 2006).

The situation is similar for shiftworkers of offshore oil installations, who live in the same environment during both work and leisure time and stay away from home for several weeks, usually working in alternating 12-hour shift schedules (6:00–18:00, 18:00–6:00). In addition, in this occupational setting, the (mal)adjustment of the circadian rhythm may be more or less pronounced and depends on whether the fast or slow rotation is adopted, job characteristics (drilling, maintenance), and working organization (Barnes et al., 1998).

Similar problems can be faced by long-haul truck and train drivers (i.e. coast-to-coast journeys, relay work), in which shiftwork, long working hours and time zone crossing interact in causing circadian disruption of the sleep/wake cycle and biological rhythms, as well as sleep deprivation, and overall fatigue (Jay et al., 2006).

SHIFTWORK 577

1.4 Biomarkers of circadian regulation and dysregulation

The production and release of nearly all hormones exhibits a circadian timing patterned on approximately a 24-hour cycle (Pandi-Perumal et al., 2007). Agents that disrupt the circadian rhythm may therefore alter hormone levels.

At present, there is no known biomarker of exposure to shiftwork, which is thought to affect circadian regulation. In the past, core body temperature or blood cortisol levels have been used as markers of circadian regulation. However, given the importance of melatonin in the regulation of circadian rhythm, an indicator of melatonin levels is considered a preferable biomarker of circadian regulation and dysregulation, and has been more commonly used in recent studies on the effects of shiftwork in humans. Melatonin levels are comparatively robust in the presence of various external influences (Pandi-Perumal et al., 2007). For example, excessive carbohydrate intake can significantly affect core body temperature and heart rate, whereas melatonin concentration remains relatively unaffected by this factor (Kräuchi et al., 2002). Furthermore, the onset of melatonin production is largely unaffected by biochemical and physiological factors, which further suggests its greater reliability to measure circadian phase position (Lewy, 1999; Lewy et al., 1999).

1.4.1 Methods of measuring circulating melatonin

(a) Melatonin in serum and plasma Plasma melatonin, which has a very short biological half-life and is rapidly

metabolized by the liver, reflects the amount of melatonin circulating at the point in time of the sample collection. Thus, measurement of melatonin in plasma at regular intervals (e.g. hourly) will map out a circadian rhythm, enabling identification of the onset of melatonin secretion, the duration of melatonin secretion, peak levels of circulating melatonin, and the time at which peak secretion occurred, and the total amount of melatonin secreted. Although such detailed information may be very useful in identifying the characteristics of the circadian rhythm in an individual, such measurement is possible only in a controlled setting (e.g. sleep laboratory), and is impractical in other applications such as an epidemiological study or other widespread population use.

(b) Melatonin in saliva Melatonin can also be measured in saliva, using several different laboratory

techniques. Salivary testing is a useful method for measuring melatonin in epidemiological studies, given that it is relatively non-invasive and generally acceptable to study participants. With proper training, study subjects can collect their own samples at home, to be later delivered to the laboratory for assay. Several researchers have found a high correlation between serum and salivary melatonin concentrations, and have concluded that salivary melatonin concentrations are reliable indices of serum melatonin


concentrations (Arendt et al., 1985; Laakso et al., 1990; Klante et al., 1997; Davis et al., 2001; Gooneratne et al., 2003). However, Laakso et al. (1990) compared salivary and serum melatonin levels and found that saliva and serum measurements were not highly correlated in individuals with low serum melatonin levels, and that the proportion of melatonin found in saliva decreased with increasing serum melatonin levels. They concluded that melatonin concentrations measured in saliva do not always consistently reflect the absolute concentrations in blood. Gooneratne et al. (2003) reported similar results in that serum and saliva melatonin levels were less correlated in individuals with low serum melatonin levels. The primary drawback to measuring melatonin in saliva is that, similar to plasma and serum measurements, salivary melatonin reflects the amount of melatonin circulating in the body at a given time-point. To capture details of the rhythm of melatonin secretion, such as the time of onset, peak levels, and cumulative secretion, one has to collect the subject’s saliva samples at regular intervals throughout the night.

(c) Melatonin in urine Arendt et al. (1985) suggested that measurement of the primary metabolite of

melatonin excreted in urine would allow the non-invasive study of pineal function, useful in a broad range of applications. If appropriately executed, measurement of melatonin in the urine reflects the cumulative amount of circulating melatonin corresponding to the time period between the prior urine void and the collection of the subsequent urine sample. Using this approach to quantify melatonin levels in urine is typically accomplished through the measurement of 6-hydroxymelatonin sulfate (aMT6s), the primary metabolite in urine, although some studies directly measure urinary melatonin. The principal methods for determination of urinary aMT6s include assay by either radioimmunoassay (RIA) or enzyme-linked immunosorbant assay (ELISA); commercial kits are available for both methods. Concentrations of aMT6s are often adjusted by urinary creatinine concentrations to account for differing urine output volume from one individual to the next, and for separate urine collections within individuals (Klante et al., 1997).

The stability of such measurements further promotes the usefulness of this technique, since long-term levels of hormones are often of interest in diseases with long latency periods. Davis et al. (2001) evaluated nocturnal aMT6s levels in a group of women 20–74 years of age over 3 consecutive days, then repeated the measurement protocol 3–6 months later. Urinary aMT6s concentrations have been shown to be highly and significantly correlated on consecutive days, as well as between measurements sessions over long time period until 5-year time period in several studies (Levallois et al., 2001; Travis et al., 2003). Levallois et al. (2001) measured urinary aMT6s concentrations over 2 consecutive days and a found similarly high correlation.

SHIFTWORK 579

(d) Comparison between blood and urinary melatonin levels

As melatonin is secreted primarily at night, studies have focused on nocturnal samples when evaluating the correlation between melatonin levels in blood and urine, and found a high degree of correlation between nocturnal measurements of urinary melatonin or urinary aMT6s, and plasma or serum melatonin. Graham et al. (1998) found a significant relationship between total nocturnal plasma melatonin and both urinary aMT6s corrected for creatinine and urinary melatonin. Combining the two urinary measures of aMT6s and melatonin accounted for 72% of the variance in total plasma melatonin. Furthermore, peak nocturnal levels of plasma melatonin were significantly related to morning levels of urinary melatonin and aMT6s. Cook et al. (2000) assessed the differences in melatonin levels between blood and urine samples collected in a laboratory-based setting with nocturnal urine samples collected in a field study, and found very high correlations (P < 0.001) between first morning void melatonin and creatinine-corrected aMT6s and both total nocturnal plasma melatonin output and peak nocturnal plasma melatonin.

Similarly high correlations have been found in studies that compared melatonin in plasma and serum with urinary melatonin and/or urinary aMT6s over a 24-hour period (Markey et al., 1985; Baskett et al., 1998). Bojkowski et al. (1987) found that total 24-hour urinary excretion of aMT6s was significantly correlated with the area under the curve of the respective profiles for plasma melatonin (r = 0.75), and plasma aMT6s (r = 0.70).

In conclusion, both urinary melatonin and urinary aMT6s are good indicators of melatonin secretion in blood with a significantly smaller variation for the former molecule (Pääkkönen et al., 2006). Such measurements in urine samples would provide a suitable tool in epidemiological settings to study the modulation of the circadian rhythm in shiftworkers.

1.5 Regulations on shiftwork

Some international directives have been issued in the last decades addressing the need for a careful organization of shift and night work and the protection of shiftworkers’ health: in particular, the International Labour Office (ILO) “Code of practice on working time” (1995) and Convention no. 171 (C171) on “Night work” (1990), and the European Directive no. 93/104/EC “concerning certain aspects of the organization of working time” (1993), which in European countries has been implemented through national legislation.


1.5.1 ILO Night Work Convention and Recommendation

(a) General population The ILO C171 Night Work Convention (International Labour Organization, 1990a)

refers only to night work, that is “all work which is performed during a period of not less than seven consecutive hours, including the interval from midnight to 5am,” and night worker, who is “an employed person whose work requires performance of a substantial number of hours of night work which exceeds a specified limit, fixed by the competent authority. This convention applies to all employed persons except those employed in agriculture, stock raising, fishing, maritime transport and inland navigation.”

In addition, the ILO R178 Night Work Recommendation (International Labour Organization, 1990b), supplementing the Night Work Convention C171, points out the following:

“Normal hours of work for night workers should not exceed eight in any 24-hour period in which they perform night work, except in the case of work which includes substantial periods of mere attendance or stand-by, in cases in which alternative working schedules give workers at least equivalent protection over different periods or in cases of exceptional circumstances recognized by collective agreements or failing that by the competent authority.

The normal hours of work of night workers should generally be less on average than and, in any case, not exceed on average those of workers performing the same work to the same requirements by day in the branch of activity or the undertaking concerned.

In occupations involving special hazards or heavy physical or mental strain, no overtime should be performed by night workers before or after a daily period of work which includes night work, except in cases of force majeure or of actual or imminent accident.

Where shift work involves night work: (a) in no case should two consecutive full-time shifts be performed, except in cases of force majeure or of actual or imminent accident; (b) a rest period of at least 11 hours between two shifts should be guaranteed as far as is possible.”

(b) Women during pregnancy and around childbirth At any point during pregnancy, once this is known, women night workers who so

request should be assigned to day work, as far as is practical. Measures shall be taken to ensure that an alternative to night work is available to

women workers who would otherwise be called upon to perform such work: (a) before and after childbirth, for a period of at least 16 weeks of which at least 8 weeks shall be before the expected date of childbirth; (b) for additional periods in respect of which a medical certificate is produced stating that it is necessary for the health of the mother or child: (i) during pregnancy; (ii) during a specified time beyond the period after childbirth fixed pursuant to subparagraph (a) above, the length of which shall be determined by the

SHIFTWORK 581

competent authority after consulting the most representative organizations of employers and workers. These measures may include transfer to day work where this is possible, the provision of social security benefits or an extension of maternity leave. During those periods, a woman worker shall not be dismissed or given notice of dismissal, except for justifiable reasons not connected with pregnancy or childbirth, and shall not lose the benefits regarding status, income, seniority and access to promotion which may attach to her regular night work position (ILO C171, 1990).

(c) Young people With regard to young people, following the first Night Work of Young Persons

(Industry) Convention (1919), the ILO Night Work of Young Persons (Industry) Convention (Revised) (1948), stated that: “young persons under eighteen years of age shall not be employed or work during the night in any public or private industrial undertaking (i.e. mines, quarries, manufactures, construction, transports, electrical-gas works, etc.). “Night” means a period of at least twelve consecutive hours. In the case of young persons under sixteen years of age, this period shall include the interval between ten o’clock in the evening and six o’clock in the morning. Moreover, in the case of young persons who have reached the age of sixteen years but are under the age of eighteen years, this period shall include an interval prescribed by the competent authority of at least seven consecutive hours falling between ten o’clock in the evening and seven o’clock in the morning. For purposes of apprenticeship or vocational training in specified industries or occupations which are required to be carried on continuously, the Convention stated that the competent authority may, after consultation with the employers’ and workers’ organizations concerned, authorise the employment in night work of young persons who have reached the age of sixteen years but are under the age of eighteen years.”

(d) Seafarers For specific groups of workers, the ILO Convention No. 180 “Concerning Seafarers’

Hours of Work and the Manning of Ships” (1996) states limits on hours of work or rest, in particular: “a) maximum hours of work shall not exceed 14 hours in any 24-hour period, and 72 hours in any seven-day period; b) minimum hours of rest shall not be less than ten hours in any 24-hour period, and 77 hours in any seven-day period; c) hours of rest may be divided into no more than two periods, one of which shall be at least six hours in length, and the interval between consecutive periods of rest shall not exceed 14 hours. Moreover, no seafarer under 18 years of age shall work at night (which means a period of at least nine consecutive hours, including the interval from midnight to five a.m.).”

(e) Long-distance drivers According to the US Bureau of Labor Statistics (2007) long-distance drivers may

drive for 11 hours and work for up to 14 hours – including driving and non-driving duties – after having 10 hours off-duty. Moreover, they may not drive after having worked for


60 hours in the past 7 days or 70 hours in the past 8 days unless they have taken at least 34 consecutive hours off-duty.

(f) Airline pilots According to the National Aeronautics and Space Administration guidelines

(Dinges et al. 1996), for standard operations including day and night flying, the duty period for air pilots should not exceed 10 hours within a 24-hour period; in case of extended flight duty periods, the limit should be fixed at 12 hours, and accompanied by additional restrictions and compensatory off-duty periods. It is also recommended that in any 7-day period, there be no extended flight duty period that encroaches on any portion of the window of circadian low (i.e. period between 2–6 am for an individual’s normal day–wake/night–sleep schedule).

Because of Federal Aviation Administration regulations, airline pilots flying large aircraft, cannot fly more than 100 hours a month or more than 1000 hours a year. Most airline pilots fly an average of 75 hours a month and work an additional 75 hours a month performing non-flying duties. To guard against pilot fatigue, which could result in unsafe flying conditions, the Federal Aviation Administration requires airlines to allow pilots at least 8 hours of uninterrupted rest in the 24 hours before finishing their flight duty.

Many countries in the world have national laws regulating night work according to ILO recommendations, whereas in many others this topic is regulated by means of collective or local agreements between parties (International Labour Organization, 1995).

1.5.2 European Directive on Working Time

(a) General population In Europe, the EU Council Directive No 93/104/EC (European Council Directive,

1993) “concerning certain aspects of the organization of working time” (re-confirmed by EU Directive 2003/88/EC):

– defined “night time” as “any period of not less than seven hours, as defined by national law, and which must include in any case the period between midnight and 5 am”; and “night worker” as (a) any worker who, during night time, works at least three hours of his/her daily working time as a normal course, and (b) any worker who is likely during night time to work a certain proportion of his/her annual working time, as defined at the choice of the Member State concerned either by national legislation or by collective agreements. On the other hand, shift work means “any method of organising work in shifts whereby workers succeed each other at the same work stations according to a certain pattern, including a rotating pattern, and which may be continuous or discontinuous, entailing the need for workers to work at different times over a given period of days or weeks; consequently, “shift worker shall mean any worker whose work schedule is part of shift work.”

SHIFTWORK 583

– forced Member States to take the measures necessary to ensure that: normal hours of work for night workers do not exceed an average of 8 hours in any 24-hour period for normal work activities, but not more than 8 hours in any 24-hour period in case of work involving special hazards or heavy physical or mental strain; every worker is entitled to a minimum daily rest period of 11 consecutive hours per 24-hour period; where the working day is longer than 6 hours, every worker is entitled to a rest break; per each seven-day period, every worker is entitled to a minimum uninterrupted rest period of 24 hours plus the 11 hours daily rest; and it should preferably include Sunday; the average working time for each seven-day period, including overtime, does not exceed 48 hours; every worker is entitled to paid annual leave of at least four weeks in accordance with the conditions for entitlement to, and granting of, such leave laid down by national legislation and/or practice; the minimum period of paid annual leave may not be replaced by an allowance in lieu, except where the employment relationship is terminated.

Implementing such directive at national level, some European countries added the quantitative criterium of 80 night shifts worked per years as minimum level for establishing the compulsory periodical medical surveillance for night workers: this limit appears as a mere technical compromise among social parties (i.e. one third of the total working days), being not supported by any evidence based on the scientific literature.

There are also some differences among countries in the definition of both “night work” and “night worker” (see Table 1.7).

Table 1.7. Legislation on night work in 15 EUa countries

Country Max. length of

night work in hours

Legislation

AUSTRIA – Nachtschwerarbeitsgesetz nr. 354/1981 (rev. 1993)– “Night work”: period of at least 6 hours between 22:00 and 06:00 for at least six nights a month. Additional breaks: 10 min paid break during the night shift. Additional vacations: 60 nightshifts per year, 2 work days, after 5 years on shift, 4 work days, after 15 years on shift, 6 work days. Health service, possibility of early retirement.

BELGIUM 8 Loi du 17/02/1997 et Loi du 04/12/1998: “Night time”: a period, generally of 8 hours, between 20:00 and 06:00. “Night work”: in principle, prohibited, but various derogations are possible.

DENMARK – The notions of night time and night worker have been defined generally in collective agreements.


Table 1.7 (contd)


night work in hours

Legislation

FINLAND – Working Hours Act 605/1996: “Night work”: work of at least 3 hours between 23.00 and 06.00. An employer must notify the labour protection authorities of regular night work, when the said authorities so request.

FRANCE – Loi 461/1998: “Night time”: period between 22:00 and 05:00 or whichever night work period between midnight and 05:00. “Night workers”: any employee working usually at least 2 times per week at least 3 hours on the period defined as night work.

GERMANY 8/10 Arbeitszeitgesetz 1994: “Night time”: a period which includes the time between 23.00 and 06.00, in the case of bakers between 22.00 and 05.00. “Night work”: every kind of work which includes more than 2 hours of night time. The working time of a night worker and shiftworker shall not exceed 8 hours, or 10 hours if within a month or a 4-weeks period where the average working hours are 8 hours per day. The night workers are entitled to a health assessment before they take up the assignment and after that, every 3 years. After the age of 50, the time is reduced to 1 year. “Night worker”: a worker who works at least 2 hours during night time. “Night workers” are those workers who usually work nights in rotating shifts system or works at night on not less than 48 days during a year. The working time of a night worker and shiftworker shall be laid out according to evidence based knowledge about human centred design of working hours from ergonomics.

GREECE 8 Presidential Decree no. 88/1999: “Night time”: period of 8 hours which includes the period between 22:00 and 06:00. “Night worker”: a worker who during night time works at least 3 hours of his daily working time or a worker who has to perform night work for at least 726 hours of his annual working time.

IRELAND 9 Statutory Instruments no. 485/1998: “Night time”: period between midnight and 07.00. “Night worker”: a) an employee who normally works at least 3 hours of his or her daily working time during night time; b) an employee whose working hours during night time, in each year, equals or exceeds 50 per cent of the total number of hours worked during the year.

SHIFTWORK 585

Table 1.7 (contd)


night work in hours

Legislation

ITALY – D.Lgs. 66/2003: “Night work”: the activity carried out in a period of at least 7 consecutive hours comprising the interval between midnight and 05.00 in the morning. “Night worker”: a) any worker who during the night period carries out, in a not exceptional way, at least 3 hours of his daily working time; b) any worker who carries out, during the night, at least a part of his normal working hours. Night work does not have to be done obligatorily by: a) the working mother of a child under 3 years of age or, alternatively, by the cohabiting father; b) the worker who is the only entrusted parent of a cohabiting child of less than 12 years of age; c) the worker who takes care of a disabled subject. Women are forbidden to work from 24.00 to 06.00, from the assessment of state of pregnancy until the first year of age of the child. Thereafter their assignment to night work is on voluntary basis until the third year of age of the child.

LUXEMBOURG – There is no general legislation on night work or night worker.

NETHERLANDS – Wet van 23/11/1995: “Night work”: work which covers all or part of the period from midnight to 06:00.

PORTUGAL 8 Decreto Lei 259/98: “Night time”: a period between 20:00 and 07:00 L.73/98: “Night work”: shall not exceed 8 hours. The night workers with risks shall not work more than 8 hours in a period of 24 hours. The employer ensures the worker the opportunity of a free health assessment before he takes up the assignment and during the period of work.

SPAIN 8 Real Decreto Lei 1/1995: “Night time”: the period which includes the interval between 22.00 and 06.00. “Night work”: shall not exceed the 8 hours in a work period of 15 days. The employer, who usually utilizes night work, has to inform the authority. “Night worker”: the worker who at night carries out at least 3 hours of its daily working time”.


Table 1.7 (contd)


night work in hours

Legislation

SWEDEN – Working Hours Act 1982: All employees shall be afforded free time for nightly rest. Such free time shall include the hours between midnight and 05:00. Exception could be made depending on the nature of the work. “Night worker”: a worker that works at least 3 hours of his daily work during night time, or a worker that most likely will work at least 38% of his annual work during the night.

UK 8 Statutory Instruments.1833/1998: “Night time”: a period the duration of which is not less than 7 hours, and which includes the period between midnight and 05:00. A nightworker’s normal hours of work, in any reference period which is applicable in his case, shall not exceed an average of 8 hours for each 24 hours. “Night worker”: a worker who, as a normal course, works at least 3 hours of his daily working time during night time, or who is likely, during night time, to work at least such proportion of his annual working time as may be specified for the purposes of these regulations in a collective agreement or a workforce agreement. An employer shall not assign an adult worker to work which is to be undertaken during periods such that the worker will become a night worker unless the employer has ensured that the worker will have the opportunity of a free health assessment before he takes up the assignment; or the worker had a health assessment before being assigned to work to be undertaken during such periods on an earlier occasion, and the employer has no reason to believe that that assessment is no longer valid.

a Council Directive 93/104/EC of 23 November 1993 concerning certain aspects of the organization of working time. Compiled by the Working Group

(b) Women during pregnancy and around childbirth For women, the EU Council Directive 92/85/EEC (European Council Directive,

1992), “on the introduction of measures to encourage improvements in the safety and health at work of pregnant workers and workers who have recently given birth or are breastfeeding,” forced Member States to take the necessary measures to ensure that such workers are not obliged to perform night work during their pregnancy and for a period following childbirth which shall be determined by the national authority competent for safety and health. These measures must entail the possibility, in accordance with national

SHIFTWORK 587

legislation and/or national practice, of transfer to daytime work, or leave from work or extension of maternity leave where such a transfer is not technically and/or objectively feasible.

In most legislations of European countries, women are prohibited to work at night from the assessment of state of pregnancy until the first year of age of the child. Thereafter, in many cases, assignment to night work is on voluntary basis until the third year of the child.

(c) Young people For young people, the European Council Directive 94/33/EC (1994) on the protection

of young people at work states that: “Member States shall adopt the measures necessary to prohibit work by children (less than 15 years of age) between 8 pm and 6 am (in case of cultural or similar activities allowed to children), and by adolescents (15–18 years of age) either between 10 pm and 6 am or between 11 pm and 7 am. For adolescents, there may be some exceptions in specific areas provided that they are supervised by an adult, but work between midnight and 4 am continues to be prohibited.

1.5.3 Scientific guidelines

The main indications for the design of better shift systems according to ergonomic criteria are (Knauth, 1996; Knauth and Hornberger, 2003; Wedderburn, 1994):

a) Quickly rotating shift systems are better than slowly rotating ones. b) Clockwise rotation (morning/afternoon/night) is preferable to counter-

clockwise (afternoon/morning/night). c) Early starts for the morning shift should be avoided. d) Prolonged work shifts (9–12 hour) should only be considered when the

workload is suitable, there are adequate breaks, and the shift system is designed to minimize accumulation of fatigue and exposure to toxic substances.

e) Shift systems should be regular and able to guarantee as many free weekends as possible.

f) Permanent night work can be acceptable only for particular working situations which require a complete adjustment to night work to guarantee the highest levels of safety. Be aware that such complete adjustment requires people to maintain the inverted sleep/wake cycle also on rest days and to avoid exposure to bright light after night shifts (i.e. wearing dark sun glasses while commuting home).

g) Adequate time off between shifts should be allowed to compensate for fatigue and sleep as quickly as possible (i.e. two shifts in the same day must be avoided), and rest days should come preferably after the night duty period to allow prompt recovery from sleep deficit and an easier return to the normal sleep/wake cycle.

h) Some flexibility in working times is desirable to give the workers the possibility of combining better work duties with family and social life.


1.6 References Arendt J, Bojkowski C, Franey C et al. (1985). Immunoassay of 6-hydroxymelatonin sulfate in

human plasma and urine: abolition of the urinary 24-hour rhythm with atenolol. J Clin Endocrinol Metab, 60:1166–1173 doi:10.1210/jcem-60-6-1166. PMID:3998065

Ariznavarreta C, Cardinali DP, Villanúa MA et al. (2002). Circadian rhythms in airline pilots submitted to long-haul transmeridian flights. Aviat Space Environ Med, 73:445–455. PMID:12014603

Barnes RG, Deacon SJ, Forbes MJ, Arendt J (1998). Adaptation of the 6-sulphatoxymelatonin rhythm in shiftworkers on offshore oil installations during a 2-week 12-h night shift. Neurosci Lett, 241:9–12 doi:10.1016/S0304-3940(97)00965-8. PMID:9502203

Baskett JJ, Cockrem JF, Antunovich TA (1998). Sulphatoxymelatonin excretion in older people: relationship to plasma melatonin and renal function. J Pineal Res, 24:58–61 doi:10.1111/j.1600-079X.1998.tb00366.x. PMID:9468119

Bøggild H, Knutsson A (1999). Shift work, risk factors and cardiovascular disease. Scand J Work Environ Health, 25:85–99. PMID:10360463

Bojkowski CJ, Arendt J, Shih MC, Markey SP (1987). Melatonin secretion in humans assessed by measuring its metabolite, 6-sulfatoxymelatonin. Clin Chem, 33:1343–1348. PMID:3608151

Cook MR, Graham C, Kavet R et al. (2000). Morning urinary assessment of nocturnal melatonin secretion in older women. J Pineal Res, 28:41–47 doi:10.1034/j.1600-079x.2000.280106.x. PMID:10626600

Costa G (1996). The impact of shift and night work on health. Appl Ergon, 27:9–16 doi:10.1016/0003-6870(95)00047-X. PMID:15676307

Costa G (2003). Factors influencing health and tolerance to shift work. Theor Issues Ergon Sci, 4:263–288 doi:10.1080/14639220210158880.

Costa G, Lievore F, Casaletti G et al. (1989). Circadian characteristics influencing interindividual differences in tolerance and adjustment to shiftwork. Ergonomics, 32:373–385 doi:10.1080/00140138908966104. PMID:2753014

Costa G, Akerstedt T, Nachreiner F et al. (2004). Flexible working hours, health, and well-being in Europe: some considerations from a SALTSA project. Chronobiol Int, 21:831–844 doi:10.1081/CBI-200035935. PMID:15646231

Davis S, Kaune WT, Mirick DK et al. (2001). Residential magnetic fields, light-at-night, and nocturnal urinary 6-sulfatoxymelatonin concentration in women. Am J Epidemiol, 154:591–600 doi:10.1093/aje/154.7.591. PMID:11581092

Dinges DF, Graeber RC, Rosekind MR et al. (1996) Principles and guidelines for duty and rest scheduling in commerical aviation. NASA Technical Memorandum 110404 AMES Research Center Moffett Field CA, May 1996.

Eriksen CA, Gillberg M, Vestergren P (2006). Sleepiness and sleep in a simulated “six hours on/six hours off” sea watch system. Chronobiol Int, 23:1193–1202 doi:10.1080/07420520601057981. PMID:17190705

European Council Directive (1992). No 92/85/EEC on the introduction of measures to encourage improvements in the safety and health at work of pregnant workers and workers who have recently given birth or are breastfeeding (tenth individual Directive within the meaning of Article 16 (1) of Directive 89/391/EEC). Off J Eur Commun, L348:1–8.

European Council Directive (1993) No 93/104/EC concerning Certain Aspects of the Organization of Working Time. Off J Eur Commun, L307: 18–24.

SHIFTWORK 589

European Council Directive (1994) No 94/33/EC L 216 on the protections of young people at work. 24 June 1994. Off J Eur Commun, L216: 12–20.

European Council Directive (2003) No 2003/88/EEC of the European Parliament and the council of 4 November 2003 concerning certain aspects of the organisation of the working time. Off J Eur Commun, L299: 9–19.

European Foundation (2007) Fourth European Working Conditions Survey. - (European Foundation for the Improvement of Living and Working Conditions). Loughlinstown, Dublin, Ireland. www.eurofound.europa.eu

Gander PH, Myhre G, Graeber RC et al. (1989). Adjustment of sleep and the circadian temperature rhythm after flights across nine time zones. Aviat Space Environ Med, 60:733–743. PMID:2775129

Gooneratne NS, Metlay JP, Guo W et al. (2003). The validity and feasibility of saliva melatonin assessment in the elderly. J Pineal Res, 34:88–94 doi:10.1034/j.1600-079X.2003.02945.x. PMID:12562499

Graham C, Cook MR, Kavet R et al. (1998). Prediction of nocturnal plasma melatonin from morning urinary measures. J Pineal Res, 24:230–238 doi:10.1111/j.1600-079X.1998.tb00538.x. PMID:9572533

IAOPA (2007). International Council of Aircraft Owner and Pilot. http://www.iaopa.org/welcome/ International Labour Organization (1948). Night Work of Young Persons (Industry) Convention

(Revised).C90 Geneva, ILO. http://www.ilo.org/ilolex/cgi-lex/convde.pl?C090 International Labour Organization (1990b). Night Work Recommendation. R178, Geneva, ILO. International Labour Organization (1990a). Night Work Convention. C171, Geneva, ILO. International Labour Organization (1995). Working time around the world. Conditions of Work

Digest. - (International Labour Office). Vol 14, Geneva, ILO. http://www.ilo.org/ilolex/english/index.htm

International Labour Organization (1996). Seafarers' Hours of Work and the Manning of Ships Convention.ILO convention No 180, Geneva, 22 October 1996.

International Labour Organization (2006). LABORSTA Internet. Yearly statistics. -.Geneva ILO http://laborsta.ilo.org/

Jansen B, Kroon H (1995). Rota-risk-profile-analysis. Work Stress, 9:245–255 doi:10.1080/02678379508256560.

Jay SM, Dawson D, Lamond N (2006). Train drivers’ sleep quality and quantity during extended relay operations. Chronobiol Int, 23:1241–1252 doi:10.1080/07420520601083409. PMID:17190709

Karlsson B, Knutsson A, Lindahl B (2001). Is there an association between shift work and having a metabolic syndrome? Results from a population based study of 27,485 people. Occup Environ Med, 58:747–752 doi:10.1136/oem.58.11.747. PMID:11600731

Klante G, Brinschwitz T, Secci K et al. (1997). Creatinine is an appropriate reference for urinary sulphatoxymelatonin of laboratory animals and humans. J Pineal Res, 23:191–197 doi:10.1111/j.1600-079X.1997.tb00354.x. PMID:9462851

Knauth P (1996). Designing better shift systems. Appl Ergon, 27:39–44 doi:10.1016/0003-6870(95)00044-5. PMID:15676310

Knauth P, Hornberger S (2003). Preventive and compensatory measures for shift workers. Occup Med (Lond), 53:109–116 doi:10.1093/occmed/kqg049. PMID:12637595


Knutsson A, Andersson H, Berglund U (1990). Serum lipoproteins in day and shift workers: a prospective study. Br J Ind Med, 47:132–134. PMID:2310717

Knutsson A (2003). Health disorders of shift workers. Occup Med (Lond), 53:103–108 doi:10.1093/occmed/kqg048. PMID:12637594

Kräuchi K, Cajochen C, Werth E, Wirz-Justice A (2002). Alteration of internal circadian phase relationships after morning versus evening carbohydrate-rich meals in humans. J Biol Rhythms, 17:364–376. PMID:12164252

Laakso ML, Porkka-Heiskanen T, Alila A et al. (1990). Correlation between salivary and serum melatonin: dependence on serum melatonin levels. J Pineal Res, 9:39–50. PMID:2231272 doi:10.1111/j.1600-079X.1990.tb00692.x

Lennernäs MA, Hambraeus L, Akerstedt T (1993). Nutrition and shiftwork: the use of meal classification as a new tool for qualitative/quantitative evaluation of dietary intake in shiftworkers. Ergonomics, 36:247–254 doi:10.1080/00140139308967879. PMID:8440220

Levallois P, Dumont M, Touitou Y et al. (2001). Effects of electric and magnetic fields from high-power lines on female urinary excretion of 6-sulfatoxymelatonin. Am J Epidemiol, 154:601–609 doi:10.1093/aje/154.7.601. PMID:11581093

Lewy AJ (1999). The dim light melatonin onset, melatonin assays and biological rhythm research in humans. Biol Signals Recept, 8:79–83 doi:10.1159/000014573. PMID:10085467

Lewy AJ, Cutler NL, Sack RL (1999). The endogenous melatonin profile as a marker for circadian phase position. J Biol Rhythms, 14:227–236 doi:10.1177/074873099129000641. PMID:10452335

Markey SP, Higa S, Shih M et al. (1985). The correlation between human plasma melatonin levels and urinary 6-hydroxymelatonin excretion. Clin Chim Acta, 150:221–225 doi:10.1016/0009-8981(85)90247-5. PMID:4064329

Pääkkönen T, Mäkinen TM, Leppäluoto J et al. (2006). Urinary melatonin: a noninvasive method to follow human pineal function as studied in three experimental conditions. J Pineal Res, 40:110–115 doi:10.1111/j.1600-079X.2005.00300.x. PMID:16441547

Pandi-Perumal SR, Smits M, Spence W et al. (2007). Dim light melatonin onset (DLMO): a tool for the analysis of circadian phase in human sleep and chronobiological disorders. Prog Neuropsychopharmacol Biol Psychiatry, 31:1–11 doi:10.1016/j.pnpbp.2006.06.020. PMID:16884842

Reinberg A, Migraine C, Apfelbaum M et al. (1979). Circadian and ultradian rhythms in the feeding behaviour and nutrient intakes of oil refinery operators with shift-work every 3--4 days. Diabete Metab, 5:33–41. PMID:446831

Romon M, Beuscart R, Frimat P et al. (1986). [Caloric intake and weight gain according to the shift schedule of shift workers]. Rev Epidemiol Sante Publique, 34:324–331. PMID:3823526

Suvanto S, Partinen M, Härmä M, Ilmarinen J (1990). Flight attendants’ desynchronosis after rapid time zone changes. Aviat Space Environ Med, 61:543–547. PMID:2369394

Tenkanen L, Sjöblom T, Härmä M (1998). Joint effect of shift work and adverse life-style factors on the risk of coronary heart disease. Scand J Work Environ Health, 24:351–357. PMID:9869306

Travis RC, Allen NE, Peeters PH et al. (2003). Reproducibility over 5 years of measurements of 6-sulphatoxymelatonin in urine samples from postmenopausal women. Cancer Epidemiol Biomarkers Prev, 12:806–808. PMID:12917214

US Bureau of Labor Statistics (2005). Occupational Outlook Handbook. Department of Labor. http://www.bls.org

SHIFTWORK 591

US Bureau of Labor Statistics (2007), Occupational Outlook Handbook, 2006–07 Edition Aircraft Pilots and Flight Engineers. Department of Labor. www.bls.gov/oco/ocos107.htm

van Amelsvoort LG, Jansen NW, Kant I (2006). Smoking among shift workers: More than a confounding factor. Chronobiol Int, 23:1105–1113 doi:10.1080/07420520601089539. PMID:17190698

Waterhouse JM, Folkard S, Minors DS (1992) Shiftwork, health and safety. An overview of the scientific literature 1978–1990. London: HMSO.

Wedderburn A (1994) Instruments for designing, implementing and assessing working time arrangements. Bulletin of European Studies on Time, 7.

Wegmann HM, Klein KE (1985) Jet-lag and aircrew scheduling. In: Folkard S and Monk TH, ed., Hours of work. Temporal factors in work scheduling., Chichester, John Wiley & Sons, pp. 263–276.


2. Studies of Cancer in Humans

2.1 Introduction

Airline personnel flying over time zones are exposed to frequent disruptions of circadian rhythm, which has similarities with exposure to shiftwork. There are studies reporting cancer risk in about ten cohorts of airline cabin crew and a similar number of studies in cockpit personnel. The cabin crew cohorts support the strong evidence of significantly increased risk of breast cancer incidence found in most independent studies. Higher diagnostic activity (screening during annual health controls) may explain part of the excess when comparing with national population rates, and it should not confound internal comparisons within differently exposed subcohorts of cabin crew. Unfortunately, the studies published so far do not demonstrate precise dose–response evaluations according to the frequency of disruptions of circadian rhythm, for which the best proxy has been duration of work as flight attendant. In most studies, the excess is observed at around 10 years after first employment, and increases weakly with increasing duration. Differences in reproductive factors explain only a small fraction of the excess, while risk attributable to radiation may explain a quarter of the excess. It is unclear whether the substantial neutron component of cosmic radiation (25–50% of the effective dose but less than 5% of the absorbed dose) increases the proportion of risk attributable to radiation – this exposure can only be studied in flight crew personnel – but it is likely that there is a major part of the excess risk in breast that must be attributable to factors others than the factors listed above. Disruptions of circadian rhythm and related hormonal effects have been repeatedly mentioned as possible causal factors, and there are no data to exclude this possibility.

Prostate cancer incidence rates from the airline pilot cohorts are above the national reference levels. This excess has decreased over decades and is likely to be related to the prostate-specific antigen tests, common among pilots much earlier they became so in the general population. In the most recent follow-up reports, the SIRs among pilots have been only slightly increased. Only one study that combined cohorts of all pilots from five Nordic countries, with detailed individual level flight histories, was able to study the independent role of the long-haul flights over time zones in an internal analysis. A significant trend in risk for prostate cancer with increasing number of long-haul flights was observed, though there were only eight cases in the highest exposure category. Hence, the evidence related to the role of circadian rhythm disruptions in causing prostate cancer is weak.

SHIFTWORK 593

2.2 Shiftwork

2.2.1 Breast cancer

Eight studies reported relative risk estimates for histologically confirmed breast cancer for female night shiftworkers, with vastly differing definitions of shiftwork in each study. The characteristics of these studies are presented in Tables 2.1–2.3. Two were prospective cohort studies (Schernhammer et al., 2001; Schernhammer & Hankinson, 2005), one was a nationwide census-based cohort study (Schwartzbaum et al., 2007), three were nested case–control studies (Tynes et al., 1996; Hansen, 2001a; Lie et al., 2006), and two were retrospective case–control studies (Davis et al., 2001; O’Leary et al., 2006). All eligible studies included caucasian women; only one study (O’Leary et al., 2006) included a small proportion of Latino and African-American women (less than 10%). The majority of women studied were postmenopausal.

(a) Prospective cohort studies (Table 2.1) The two prospective cohort studies of night shiftwork and breast cancer risk used data

from the Nurses’ Health Study cohorts (NHS and NHS II) (Schernhammer et al., 2001; Schernhammer et al., 2006). The NHS began in 1976, when 121 701 registered nurses 30–55 years of age and living in 11 large US states were enrolled and completed a questionnaire comprising items about their health status, medical history, and known or suspected risk factors for cancer. Since baseline, questionnaires have been mailed biannually with the exception of lifetime history of night work in years, which was only assessed once (in 1988). Follow-up data are available for more than 90% of the ongoing cohort. In 1988, the study participants were asked how many years in total they had worked rotating night shifts with at least three nights per month, in addition to days or evenings in that month. The second cohort, NHS II, was designed in a very similar fashion. It started in 1989, when 116 671 registered female nurses (no overlap with NHS) 25–42 years of age, and from 14 US states were enrolled. Since 1989, they have completed biennial questionnaires that include items about their health status risk factors for chronic disease. Response rates to questionnaires are at 90%. In NHS II, the 1989 baseline questionnaire included detailed questions on total months during which study participants had worked on rotating night shifts for at least three nights per month in addition to days or evenings in that month. This information was updated in 1991, 1993, 1997, and 2001. Questions were asked regarding both rotating night shifts and permanent night shifts for 6 months or more in this cohort.

In the NHS, Schernhammer et al. (2001) followed a total of 78 562 women who answered the 1988 question on night work and were cancer-free at baseline over 10 years (1988–1998): of these women, 2441 incident breast cancer cases were documented during that time. The relative risks (RRs) for breast cancer associated with rotating night work compared to women who reported never having worked rotating night shifts, after

594 IA

RC

MO

NO

GR

APH

S VO

LUM

E 98

Table 2.1. Cohort studies of night shiftwork and breast cancer

Reference, location, name of study

Cohort description

Exposure assessment

Organ site (ICD code)

Exposure categories

No. of cases/deaths

OR or RR (extreme group versus referent)

Adjustment for potential confounders

Comments

Schernhammer et al. (2001) USA Nurses’ Health Study (NHS)

Prospective cohort study of 121 701 registered nurses from 11 large states, established in 1976; follow-up from 1988–1998

Self-reported life time years on rotating night shifts, one-timed assessment in 1988; rotating night shifts were defined as “at least 3 nights per month, in addition to evenings and afternoons in that month”

Breast cancer

Years of rotating night work Never 1–15 15–29 ≥30 P for trend

925 1324 134 58

1.0 (ref) 1.08 (0.99–1.18) 1.08 (0.90–1.30) 1.36 (1.0–1.78) 0.02

Age, age at menarche, parity, age at first birth, weight change, BMI, family history of breast cancer, benign breast disease, oral contraceptive use, age at menopause, alcohol consumption, use of postmenopausal hormones, menopausal status, height

SH

IFTWO

RK

595

Table 2.1 (contd)


Cohort description

Exposure assessment


Exposure categories

No. of cases/deaths



Comments

Schernhammer et al. (2006) USA Nurses’ Health Study II (NHS II)

Prospective cohort study of 116 087 registered nurses from 14 states, established in 1989; follow-up from 1989–2001

Self-reported life time years on rotating night shifts, one-timed assessment in 1989; biannual update; rotating night shifts were defined as “at least 3 nights per month, in addition to evenings and afternoons in that month”

Breast cancer

Years of rotating night work Never 1–9 10–19 20+ P for trend

441 816 80 15

1.0 0.98 (0.87–1.10) 0.91 (0.72–1.16) 1.79 (1.06–3.01) 0.65

Age, age at menarche, parity, age at first birth, BMI, family history of breast cancer, benign breast disease, alcohol consumption, oral contraceptive use, smoking status, menopausal status, age at menopause, physical activity, postmenopausal hormone use

596 IA

RC

MO

NO

GR

APH

S VO

LUM

E 98

Table 2.1 (contd)


Cohort description

Exposure assessment


Exposure categories

No. of cases/deaths



Comments

Schwartzbaum et al. (2007) Sweden Register-based – all female residents of Sweden in the work force at census in 1960 and 1970

Register-based retrospective cohort study; 1 148 661 female workers; follow-up 1971–1989; 70 breast cancer cases among 3057 women with night work (40%)

Usual occupation & work hours (three-shift schedules and others) to define occupations with a large proportion of workers with night work; from in-person interview from annual survey of living conditions (1977–1981) among 55 323 randomly invited Swedes (84% participated)

Breast cancer

Shiftwork in 1970Shiftwork in both 1960 & 1970

70 28

0.94 (0.74–1.18) 0.97 (0.67–1.40)

Age, socioeconomic status, occupational position (employed manager, other employee, self-employed with employees, self-employed without employees), county of residence (marital status and urbanization not important)

Shiftwork defined as occupations with at least 40% of the workers either reporting that they worked rotating shifts with 3 possible shifts or had work hours during the night ≥1 day before interview

SHIFTWORK 597

controlling for known breast-cancer risk factors, were as follows: for 1–14 years, 1.08 (95% CI: 0.99–1.18); for 15–29 years, 1.08 (95% CI: 0.90–1.30); and for 30 or more years, 1.36 (95% CI: 1.04–1.78). The risk increased with increasing numbers of years in shiftwork (P for trend = 0.02). [The main strengths of this study are the prospective assessment of night work information and a wide range of potential confounding factors in a well defined occupation cohort of nurses, as well as the high follow-up rate (> 90%). Limitations of this study are its one-time assessment of night work and the inclusion of permanent night workers as well as those who worked < 3 nights per month among the unexposed reference group, which may have skewed the results towards the null].

Similarly, in 115 022 predominantly premenopausal women in the NHS II, Schernhammer et al. (2006) found an elevated breast cancer risk of 1.79 (95% CI: 1.06–3.01; P = 0.65) among women who worked 20 or more years of rotating night shiftwork compared with women who reported never having worked rotating night shifts, with 1352 incident breast cancer cases accruing over 12 years of follow-up (1989–2001). [The main strengths of this study are the prospective and updated assessment of rotating night work history and a wide range of potential confounding factors in a well defined occupational cohort of nurses, as well as the high follow-up rate (90%). Limitations are the inclusion of those who worked < 3 nights per month among the unexposed reference group, and the relatively small number of women (n = 15 women) in the category with longer durations of night work].

Schwartzbaum et al. (2007) found no increase in risk in female breast cancer from their definition of night work, based on 28 observed breast cancers versus 28.91 expected, diagnosed during 1971–1989. The design is a retrospective registry-based ecological cohort study comprising all 1 148 661 Swedish women that were active in the workforce according to both 1960 and 1970 census reports. Workers were followed up for breast cancer morbidity by linkage to the Swedish Cancer Registry. Information on occupation was derived from the censuses, which included each worker’s industry and socioeconomic status. The annual surveys of living conditions (conducted during 1977–1981) among 46 438 randomly selected Swedish subjects who participated in a personal interview were used for assessing night work. Questions were asked regarding the usual occupation, work hours, and when they had started and ended working each day during the week preceding the interview. Shiftworkers were then defined as those who reported that their workplace had a rotating schedule with three or more possible shifts per day or had work hours during the night (any hour between 01:00 and 04:00) at least one day during the week preceding the interview. They classified as shiftworkers people working in job titles and industry combinations (from the censuses) with at least 40% shiftwork (as defined above). The reference group in their analyses comprised people in occupation–industry combinations in which less than 30% stated that they were shiftworkers. In analyses using 1970 census information for the definition of exposure, no increase in risk was reported among women with an occupation that was classified as shiftwork. Sub-analyses in this paper (which comprised all men and women working in Sweden) also considered 70% of shiftworkers as definition for occupation classification but due to


small sample size, this was not done for the women. [The weaknesses of this study include the implausibly small proportion of women working night shifts (only 0.3% worked in occupations with at least 40% shiftworkers working at least 20 hours per week), inadequate control for confounding, and that the three most common occupations that fell into their shiftwork classification were rather unusual (crane and hoist operators, delivery women in paper and paper-products manufacturing, printing and publishing industries, and midwives)].

(b) Nested case-control studies (Table 2.2) Tynes et al. (1996) conducted a case–control study nested within a population-based

cohort study of 2619 female Norwegian radio and telegraph operators working at sea and certified to work between 1920–1980, and followed up during 1961–1991. In total, 50 breast cancer cases were identified by linkage to the National Norwegian Cancer Registry, and each case was matched to four to seven disease-free controls from the cohort. For cases and controls, job histories on ships were collected and shiftwork as well as travel through time zones were classified for each ship mentioned in the job histories to define shiftwork. Shiftwork constituted frequent presence in the radio room both at night and during the day. After controlling for duration of employment, the SIR for breast cancer in this cohort was 1.5 (95% CI: 1.1–2.0). In the nested case–control study, there appeared to be an increased risk of breast cancer in women ≥ 50 years of age with increasing cumulative exposure to shiftwork, compared to no shiftwork (low exposure 0–3.1 years, adjusted for duration of employment, RR, 3.2, 95% CI: 0.6–17.3; high exposure 3.1–20.7 years, adjusted for duration of employment, RR, 4.3, 95% CI: 0.7–26.0; P for trend = 0.13). [The strength of this study is the use of internal controls, whereas the main limitation is its lack of control for confounding by breast cancer risk factors].

Hansen (2001a) conducted a population-based case–control study nested within the cohort of all female employees in Denmark established from the nationwide pension fund data, including information on all employments held since 1964. In total, 7035 women with incident breast cancer were identified by individual linkage to the files of the nationwide Danish Cancer Registry. Control subjects free of breast cancer were randomly drawn from the pension fund files and matched on year of birth and sex. The individual employment histories for cases and controls were reconstructed using files of the nationwide pension fund. Night work definition was based on information obtained from a nationwide interview-based survey on living and working environment conditions in 1976 among 2603 women. Trades in which at least 60% of female responders worked at night were considered to have a predominant night time schedule, whereas responders working in most trades with less than 40% reported night time schedules were regarded as day workers. The RR of breast cancer was 1.5 (95% CI: 1.3–1.7; 434 cases) among women who worked at least half a year at least 5 years before diagnosis in such trades, after controlling for age, social class, age at birth of first child, age at birth of last child, and number of children. For the subgroup of women with more than 6 years predominantly working at night, the RR was 1.7 (95% CI: 1.3–1.7; 117 cases). In further

SH

IFTWO

RK

599

Table 2.2. Nested case–control studies of night shiftwork and breast cancer


Cohort description

Exposure assessment

Exposure categories

No. of cases/deaths



Comments

Tynes et.al. (1996) Norway Telecom cohort

Cohort of 2619 female radio and telegraph operators at sea, certified between 1920–1980; follow-up from 1961–1991. The nested case–control component comprised 50 cancer registry-identified cases and 4–7 matched (year of birth) controls

Collected detailed job histories from Norwegian seamen registry; “Work at night with exposure to artificial light.” From cases and controls, detailed information on job histories on ship as well as shiftwork and travel through time zones was collected, classified by “ship”

Shiftwork in women age < 50 None <3.1 yrs. >3.1 yrs P for trend Aged 50+ None <3.1 yrs. >3.1 yrs P for trend

12 5 12 3 6 12

1.0 (ref) 0.3 (0.1–1.2) 0.9 (0.3–2.9) 0.97 1.0 (ref) 3.2 (0.6–17.3) 4.3 (0.7–26.0) 0.13

Age, duration of employment, parity, and age at first birth

600 IA

RC

MO

NO

GR

APH

S VO

LUM

E 98

Table 2.2 (contd)


Cohort description

Exposure assessment

Exposure categories

No. of cases/deaths



Comments

Hansen (2001a,b) Denmark Linkage of Nationwide registries

Nested case–controls study; 7565 cancer-registry-derived women with breast cancer, 1:1 matched controls (year of birth and sex), follow-up 1964–1999

Individual employment histories were obtained from files of national pension fund

All night work combined in trades with >60% night work Employed >6 years Nurses

434 117 –

1.5 (1.3–1.7) 1.7 (1.3–1.7) 1.3 (1.1–1.4)

Age, social class, age at birth of first child, age at birth of last child, number of children

Considered as night workers if employed ≥0.5 year in ≥1 trade in which ≥60% of the female responders had night time schedules Trades: beverage manufacture, land transport, catering, air transport

SH

IFTWO

RK

601

Table 2.2 (contd)


Cohort description

Exposure assessment

Exposure categories

No. of cases/deaths



Comments

Lie et al. (2006) Norway Cohort of Norwegian nurses

Case–control study [537 cancer-registry-identified cases and 1:4 matched (year of birth) controls] nested within the 44 835 nurses-comprising cohort of Norwegian nurses; cases occurred between 1960–1982

Total work history reconstructed from occupational information for nurses from Norwegian Board of Health’s registry, censuses 1960, 1970, & 1980

Years night work 0 1–14 15–29 30+ P for trend

50 362 101 24

1.0 (ref) 0.95 (0.67–1.33) 1.29 (0.82–2.02) 2.21 (1.10–4.45) 0.01

Total employment time as a nurse & parity; matched by birth year


sub-analyses, the RR for nurses was also evaluated, a group in which 41% were considered having predominant night work (Hansen, 2001b), and a significantly increased risk of breast cancer was found (RR, 1.3; 95% CI: 1.1–1.4). [The strength of this study is its high number of incident cases and the apparent lack of selection and information bias due to use of routine data; its limitations include the crude exposure assessment with potential for non-differential misclassification as well as incomplete adjustment for confounding, in particular alcohol drinking.]

Lie et al. (2006) conducted a nested case–control study within a cohort of 44 835 Norwegian nurses based on information from the registry of the Norwegian Board on Health,established in 1949. In total, 537 breast cancer cases diagnosed during 1960–1982 were identified by linkage with the files of the nationwide cancer registry. Four age-matched controls were selected at random from the cohort, using incidence density sampling. Reconstruction of total work history was based on the nurses’ registry (self-report of work history; until 1960 yearly updates, thereafter sporadically) and census information (1960, 1970, and 1980), accumulating from first year of employment until termination of the last employment. Based on experience, it was assumed that nurses employed at infirmaries worked nights (with the exception of managerial jobs, teaching, physiotherapy, and outpatients departments), whereas it was assumed that work sites other than infirmaries involved day work only. The authors found an association between duration of night work and breast cancer risk (P for trend = 0.01). The RR associated with > 30 years of night work was 2.21 (95% CI: 1.10–4.45), after adjustment for total employment time as a nurse and parity. [The main strength of this study is its high number of cases and the internal comparison, whereas limitations of this study are a lack of complete control for confounding as well as the potential for exposure misclassification, which is likely to be non-differential.]

(c) Case–control studies (Table 2.3) Davis et al. (2001) conducted a case–control study of 813 women with breast cancer

aged 20–75 years and 793 controls free from breast cancer. Cases were identified by the Cancer Surveillance System of Seattle, Washington, USA and controls were identified by random-digit dialling, frequency-matched on age (75% participation rate for controls). In-person interviews were performed from 1992–1995 to collect information about sleeping habits and light-at-night exposure during the 10 years before diagnosis as well as lifetime occupational history. The authors defined night work as at least one “graveyard” shift per week in the 10 years before diagnosis. “Graveyard” shiftwork was described as “beginning work after 19:00 and leaving work before 09:00”. The RR of breast cancer was 1.6 (95% CI: 1.0–2.5) among women who had ever worked “graveyard” shifts. The RR of breast cancer was 1.06 for each hour increase per week of “graveyard” shift work (P = 0.03), after controlling for parity, family history of breast cancer, oral contraceptive use, as well as recent discontinued use of hormone replacement therapy. [The strengths of this study include its attempt to accurately define shiftwork assessment. One of the main

SH

IFTWO

RK

603

Table 2.3. Case–control studies of night shiftwork and breast cancer


Cohort description

Exposure assessment


Exposure categories

No. of cases/deaths



Comments

Davis et al. (2001) Washington, USA

Cancer register based case–control study; case ascertainment (n=813) between 1992–1995, 793 matched (5-year age groups) controls identified by random-digit dialling

Information on sleeping habits, light exposure, lifetime occupational history obtained from in-person interview, considered as night workers if ≥1 graveyard shift/wk (8 hrs) in 10 years before diagnosis.

Breast cancer

Years worked ≥3 nights/wk None <1 1–3 3–4.6 4.7+ P for trend

682 19 20 9 33

1.0 (ref) 1.2 (0.6–2.3) 1.4 (0.7–2.8) 0.6 (0.3–1.5) 2.3 (1.2–4.2) 0.01

Parity, family history of breast caner (mother or sister), oral contraceptive use (ever), and recent (<5 years) discontinued use of hormone replacement therapy

Graveyard shift work defined as “beginning work after 7:00 pm and leaving work before 9:00 am.”

O’Leary et al. (2006) Long Island, New York, USA Electromagnetic Fields and Breast Cancer on Long Island Study (EBCLIS)

Case–control study with 576 registry-identified cases and 585 1:1 matched (age in 5-year age groups) population-based controls; cases occurred between 1996–1997

Occupational history since age 16 & residential light-at-night exposures (sleep hours; frequency of turning on lights during night; length of time light was on) from in-person interview

Breast cancer

Any evening or overnight shiftwork Any evening shiftwork only Any overnight shiftwork only

174 148 26

1.04 (0.79–1.38) 1.21 (0.90–1.64) 0.55 (0.32–0.94)

Age (matched by 5-year age groups); parity; education; family history of breast cancer; history of benign breast disease


limitations is the retrospective assessment of shiftwork with a modest potential for recall bias].

O’Leary et al., (2006) conducted a case–control study in Long Island, New York, USA – the Electromagnetic Fields and Breast Cancer on Long Island Study (EBCLIS). They did not observe an association between night work and breast cancer risk (any evening or overnight shiftwork versus none, OR, 1.04; 95% CI: 0.79–1.38; only overnight shiftwork, OR, 0.55; 95% CI: 0.32–0.94). This study was built into another population-based case–control study among residents of Nassau and Suffolk counties. Cases were recorded during 1996–1997. Controls were frequency-matched by age and came from two different sources: 1) controls less than 65 years old were identified by random-digit dialling; 2) controls of age 65 and above were selected from the Health Care Financing Administration rosters. To evaluate the effects of electromagnetic frequency, women from within this case–control study were selected according to their degree of residential stability (EBCLIS component). EBCLIS comprised 576 breast cancer cases and 585 matched (1:1) population-based controls. In-person interviews were held to gather information on occupational history since the age 16 years as well as residential light-at-night exposures (sleep hours; frequency of turning on lights during night; length of time light was on). Shiftwork was defined as ‘ever’ working in at least one job during the past 15 years that included evening shifts (could start in the afternoon and end as late as 02:00), overnight shifts (could start as early as 19:00 and continue until the following morning), and various combinations thereof. The reference group comprised women who reported never having had a job involving shiftwork. Results were adjusted for age (matched by 5-year age groups), parity, education, family history of breast cancer, and history of benign breast disease. [An extreme and unlikely high proportion of controls (36.9%) and cases (35.7%) reported any 'evening or overnight shiftwork'; other limitations were the retrospective assessment of exposures and that the control selection was conducted from two different sources, introducing additional potential for bias].

(d) Meta-analysis Megdal et al. (2005) conducted a meta-analysis that summarized six of the eight

studies on night work (excluding the two most recent studies that gave negative results) and breast cancer, and found an increased risk for breast cancer (RR, 1.51; 95% CI: 1.36–1.68).

(e) Studies of biomarkers for night work (urinary melatonin) and breast cancer risk (Table 2.4)

Melatonin, the main biomarker for circadian dysregulation, can be measured in the urine by 6-sulphatoxymelatonin (aMT6s), the major metabolite of melatonin.

Skene et al. (1990) compared mean urinary aMT6s levels (measured by RIA) collected from 24-hour urine samples from British women attending a breast cancer screening clinic before biopsy and 160 normal female residents of Guernsey, the United Kingdom.

SH

IFTWO

RK

605

Table 2.4. Studies of biomarkers for light exposure (urinary melatonin) and breast cancer risk

Reference, study

Country or cohort and time period under observation

Source of information for exposure (i.e., light-at-night exposure)

Definition of biomarker for light exposure

Exposure category

No. of exposed cases


Adjustment for potential cofounders

Comment

Travis et al., (2004)

Nested case-control study; 5093 female residents of Guernsey (UK) recruited into a cohort of hormones and breast cancer (Guernsey III). Questionnaire at baseline (between 1977–1985);127 incident breast cancer cases that occurred before November 1, 2001 (mean follow-up ~12.6 yrs); 353 controls subjects; (1:2 matched for age, recruitment date, menopausal status, day of menstrual cycle/years postmenopausal); premenopausal: 77 cases, 214 controls postmenopausal: 50 cases, 139 controls

24-hour urine sample collected within ~16 days of recruitment

aMT6s by RIA; adjusted for urinary creatinine; tertiles of melatonin concentrations

aMT6s ng/mg creatinine High ≥18.32 ≥12.98

46 26 20

Overall 0.99 (0.58–1.70) Premenopausal 0.99 (0.45–2.17) Postmenopausal 1.09 (0.46–2.60)

Matching factors (age ; date of recruitment, menopausal status, day of menstrual cycle that urine was collected or number of years post-menopausal); age, BMI, medication use thought to influence melatonin production, family history of breast cancer, parity and age at first birth, age at menarche, oral contraceptive use, season of urine collection, stage of menstrual cycle

606 IA

RC

MO

NO

GR

APH

S VO

LUM

E 98

Table 2.4 (contd)

Reference, study

Country or cohort and time period under observation

Source of information for exposure (i.e., light-at-night exposure)

Definition of biomarker for light exposure

Exposure category



Adjustment for potential cofounders

Comment

Schern-hammer & Hankinson (2005)

Nested case–control study; USA, Nurses Health Study II, prospective cohort study; 147 incident invasive breast cancer cases that occurred between 1996–1999 and 2001 and ≥91 matched (1:2) controls

First morning urine collection

aMT6s by ELISA; adjusted for urinary creatinine; quartiles of melatonin concentration

Urinary aMT6s ng/mg creatinine

23 Overall 0.59 (0.4–1.00) P for trend 0.13

Matching factors (year of birth, menopausal status at urine collection, month, year, time of day & luteal day of menstrual cycle at urine collection, fasting status at urine collection, ethnicity); age at menarche, parity, age at first birth, BMI, family history of breast cancer, benign breast disease, alcohol consumption, antidepressant use

Sub-analyses excluding night workers provided similar results; ~24% of the sample were postmenopausal

SHIFTWORK 607

Mean levels for aMT6s excretion were 7.8±1.1 μg/24 hours in 14 benign cases, and 4.1±0.9 μg /24 hours in ten malignant cases. However, the lower aMT6s levels in the malignant cases compared to benign cases was entirely due to the difference in age between the two groups.

Only two studies have been published to date evaluating an association between melatonin, the main biomarker for the circadian rhythm, and risk of breast cancer.

Travis et al. (2004) conducted a case–control study nested within 5093 female residents of Guernsey, the United Kingdom, which were recruited into a cohort for the study of hormones in relation to breast cancer (Guernsey III). Overall, 127 incident breast cancer cases occurred before November 2001 (mean follow-up ~12.6 years). A total of 353 control subjects were matched 1:2 to these cases. There were 77 premenopausal cases (214 controls), and 50 postmenopausal cases (139 controls). A questionnaire was distributed at baseline between 1977–1985. Urine samples (24-hour) were collected within ~16 days of recruitment, and aMT6s concentrations were measured by RIA, and adjusted for urinary creatinine. Tertiles of melatonin concentrations were created, and the overall RR was 0.99 (95% CI: 0.58–1.70, comparing highest versus lowest tertile).

Schernhammer & Hankinson (2005) also conducted a nested case–control study of similar size and with primarily premenopausal women in the USA, and nested within the NHS II, a prospective cohort study of registered nurses only. A total of 147 incident invasive breast cancer cases that occurred in this cohort were enrolled into the nested case–control study during 1996–1999 and in 2001, and 291 controls were matched (1:2) to these cases; approximately 24% of participants were postmenopausal. During 1996–1999, a first morning urine sample was collected from roughly a third of the NHS II cohort. Concentrations of aMT6s were measured by ELISA, and adjusted for urinary creatinine. Quartiles of aMT6s level were created. Overall, there was a reduction in breast cancer risk among those in the highest quartile of melatonin (RR, 0.59; 95% CI: 0.4–1.00). Sub-analyses excluding night workers provided similar results.

2.2.2 Prostate cancer

(a) Cohort studies (Table 2.5) A total of 14 052 men from 21 areas in Japan, 40–65 years old at baseline in 1988–

1990, were extracted as a subcohort of the Japan collaborative cohort study for evaluation of cancer risk (Kubo et al., 2006). A self-administered questionnaire was used to gather information at baseline on exposures related to lifestyle and work. At baseline, study participants were asked which form of work had they primarily been engaged in: daytime, fixed night or rotating shiftwork. In total, 31 cases of prostate cancer were documented from cancer registries during follow-up, based on 111 974 person–years (mean 8.0 years) from baseline until the end of 1997. Multivariate adjusted relative risks based on Cox proportional hazards models of fixed night shifts and rotating shifts were 2.3 (95% CI: 0.6–9.2; three cases), and 3.0 (95% CI: 1.2–7.7; seven cases), respectively, compared with

608 IA

RC

MO

NO

GR

APH

S VO

LUM

E 98

Table 2.5. Cohort studies


Cohort description

Exposure assessment


Exposure categories

No. of cases/ deaths

Relative risk (95% CI)


Comments

Kubo et al. (2006), Japan, Japan Collaborative Cohort

Prospective cohort of 14 052 males, aged 40–65 years old enrolled from 45 areas in Japan between 1988–1990. Information on prostate cancer was obtained from cancer registries

Self-administered questionnaires at baseline included information on type of work schedule

Prostate Day time work Fixed night Rotating shift

21 3 7

1.0 (ref) 2.3 (0.6–9.2) 3.0 (1.2–2.7)

Age, study area, family history of prostate cancer, BMI, smoking, alcohol drinking, job type, physical activity at work, workplace, perceived stress, educational level and marriage status

Schernhammer et al. (2003) USA, American Nurses Health Study

Prospective cohort of 78 586 American nurses with a baseline question on rotating night work in 1988. Follow-up for colorectal cancer was through 1998

Self-reported from postal questionnaires: Rotating night shift was defined as “at least 3 nights per months, in addition to evenings and afternoons in that month”

Colorectal Colon Rectum

No rotating night shifts 1–14 years ≥15 years No rotating night shifts 1–14 years ≥15 years No rotating night shifts 1–14 years ≥15 years

229 303 70 137 169 41 41 48 14

1.00 (ref) 1.00 (0.84-1.19) 1.35 (1.03-1.77) P for trend 0.04 1.00 (ref) 0.93 (0.74–1.17) 1.32 (0.93–1.87) P for trend 0.26 1.00 (ref) 0.86 (0.56–1.30) 1.51 (0.82–2.81) P for trend 0.15

Tobacco smoking, BMI, physical activity, aspirin use, colorectal cancer in relatives, endoscopy use, consumption of red meat, alcohol consumption, total calorie intake, postmenopausal hormones, menopausal status, height

No major differences in risk were seen for right or left colon, or colon and rectum separated

SH

IFTWO

RK

609

Table 2.5 (contd)


Cohort description

Exposure assessment


Exposure categories




Comments

Viswanathan et al. (2007), USA, American Nurses Health Study

Prospective cohort of 53 487 American nurses with a baseline question on rotating night work in 1988. Follow-up for endometrial cancer to 2004

Self-reported from postal questionnaires: Rotating night shift was defined as “≥3 nights/month, in addition to evenings & afternoons in that month”

Endometrial No rotating night shifts 1–9 yrs 10–19 yrs ≥20 yrs

210 224 43 38

1.00 (ref) 0.89 (0.74–1.08) 1.06 (0.76–1.49) 1.47 (1.03–2.10) P for trend 0.04

Age, age at menarche, age at menopause, parity, BMI, duration of oral contraceptive, postmenopausal hormones, hypertension, diabetes, and pack–years of tobacco smoking

When stratifying for obesity, women with BMI>30 and having at least 20 years with rotating shifts had a more than 2-fold significant increased risk (P for trend 0.003)

610 IA

RC

MO

NO

GR

APH

S VO

LUM

E 98

Table 2.5 (contd)


Cohort description

Exposure assessment


Exposure categories




Comments

Taylor & Pocock, (1972), England and Wales

Retrospective cohort study of 8603 manual workers all born before 1920 from 10 companies in England & Wales continuously employed for ≥10 years followed-up for death during 1956–1968.

Detailed job information since 1946, including work hours and types of shifts was obtained from payrolls kept at the companies

All cancers Lung Stomach Bladder Leukaemia

Day Shift Ex-shift* [Shift/Day] Day Shift Ex-shift Day Shift Ex-shift Day Shift Ex-shift Day Shift Ex-shift

201 219 29 [219/201] 95 94 13 33 36 4 4 7 2 6 2 2

[1.02 (0.88–1.17)] [1.16 (1.02–1.32)] [1.12 (0.75–1.61)] [1.14 (0.94–1.38] [1.09 (0.80–1.33)] [1.11 (0.90–1.36)] [1.15 (0.60–1.97)] [1.24 (0.85–1.74)] [1.43 (1.00–1.98)] [1.14 (0.31–2.93)] [0.56 (0.15–1.44)] [1.06 (0.42–2.19)] [4.00 (0.48–14.4)] [1.54 (0.57–3.35)] [0.54 (0.07–1.95)] [4.00 (0.48–14.4)]

Age, calendar time

SH

IFTWO

RK

611

Table 2.5 (contd)


Cohort description

Exposure assessment


Exposure categories




Comments

Schwartzbaum, (2007), Sweden

Ecological cohort study including male and female members of the working Swedish population participating in censuses in both 1960 and 1970, including information on usual job and industry, and followed-up for cancer in the cancer registry during 1971–1989

Job–exposure matrix based on surveys from 1977 to 1981; exposed defined as working in industries (≥20 hours/week) where at least 40% had work schedules with three or more possible shifts per day or work hours during the night at least one day per week

All cancers (29 for men and 18 for women)

Men Women

3799 103

1.01 (0.98–1.05) 1.00 (0.82–1.21)

Age, socioeconomic status, occupational position, county of residence

Significantly increased relative risks observed for kidney, skin, and other and unspecified cancers (men), whereas none of the female cancers were significantly different from unity. [Misclassification of shiftwork status is an invalidating problem]

BMI, body mass index Ex-shift, men who did not qualify as shiftworkers but had done > 6 months on shiftwork & subsequently transferred to day work. They came under observation when they had completed 10 yrs’ employment & the first 6 months of day work following their period of shiftwork. They remained under observation until the end of 1968 or until they had done a further 6 months of shiftwork.


predominantly day workers. [The major limitation of the study is the lack of statistical power, short follow-up for prostate cancer and a limited measure for shiftwork.]

(b) Case-control studies (Table 2.6)

A case–control study based on a cancer registry among residents of north-eastern Ontario, Canada, included 760 cases of prostate cancer, 45–85 years of age, and diagnosed during 1995–1998 (Conlon et al., 2007). Cases were frequency-matched by age to 1632 male controls. A comprehensive mailed questionnaire was designed to gather information on exposures to lifestyle factors, and on each job held for one or more years, including information on usual work time (daytime shift, evening/night shift, rotating shift or other). The adjusted OR for ‘ever’ having worked rotating shifts on a full-time basis was 1.19 (95% CI: 1.00–1.42, 369 cases). Analyses of the duration in years of full-time rotating shifts (P for trend = 0.05) and age working the first full-time rotating shift (P for trend = 0.03) showed significant trends, but years since first full-time shifts did not show a significant trend (P for trend = 0.16). [The Working Group noted that the proportion of cases and controls classified with rotating shiftwork seemed unrealistically high and there was a lack of statistical power.]

2.2.3 Colorectal cancer (Table 2.5)

A prospective cohort study based on the American Nurses Health study including 78 586 nurses at baseline in 1988 was used for evaluating the association between colorectal cancer risk and rotating night work (Schernhammer et al., 2003). Nurses completed a comprehensive questionnaire, including a question on how many years in total they had worked rotating night shifts at least three nights per month in addition to working days or evenings in that month. Based on 758 903 person–years during 1988–1998, a total of 602 cases of colorectal cancers were recorded. Cox proportional hazard models were used to estimate relative risks adjusted for potential confounders (tobacco smoking, body mass index, physical activity, aspirin use, colorectal cancer in relatives, endoscopy use, consumption of red meat, alcohol consumption, total caloric intake, postmenopausal hormones, menopausal status, and height). Compared with nurses who had never worked night shifts for at least three days per month, nurses who worked such shifts for 1–14 years and for at least 15 years had multivariate-adjusted RRs of 1.00 (95% CI: 0.84–1.19) and 1.35 (95% CI: 1.03–1.77), respectively. RRs adjusted for age only were similar and were reported as 1.00 (95% CI: 0.84–1.18) and 1.44 (95% CI: 1.10–1.89), respectively. Results for distinct sites such as right and left colon, combined colon, and rectum only marginally changed the results for the combined colorectal results. [Misclassification due to the relative crude definition of night shiftwork was likely to have resulted in bias towards the null.]

SH

IFTWO

RK

613

Table 2.6. Shiftwork - other sites than breast cancer. Case–control studies

Reference, study location and period


Characteristics of cases

Characteristics of controls

Exposure assessment


Exposure categories Relative risk (95% CI)*


Conlon et al. (2007), Ontario, Canada, 1995–98

Prostate 760 cancer-registry-identified cases, aged 45–84 years and diagnosed during 1995–1998

1632 controls frequency-matched on age and from the same residence

Postal questionnaire (25 pages)

369 115 87 81 86

Rotating shiftwork Years of shiftwork ≤7 7.1–22.0 22.1–34.0 >34.0 P for trend

1.19 (1.00–1.42) 1.44 (1.10–1.87) 1.14 (0.86–1.52) 0.93 (0.70–1.23) 1.30 (0.97–1.74) 0.05

Age, family history of prostate cancer


2.2.4 Endometrial cancer (Table 2.5)

Another prospective study based on the American Nurses Health study cohort included 53 487 women with an intact uterus who answered a question on rotating night work in 1988 (Viswanathan et al., 2007). They were followed-up for endometrial cancer up to mid-2004, resulting in 515 cases (720 698 person–years). The RR was 1.47 (95% CI: 1.03–2.10) for nurses with 20 or more years of rotating shiftwork. When stratifying by body mass index, the RR was 2.09 (95% CI: 1.24–3.52) in the subgroup of nurses with a body mass index >30 kg/m2 and at least 20 years of rotating shiftwork. In contrast, there was no difference in calorie consumption across night work categories. A significant trend (P = 0.003) of increasing relative risk was seen with increasing duration of rotating shiftwork in the group classified as obese. No significantly increased risk was observed in the group with a body mass index <30. The RRs were adjusted for potential confounders (age, age at menarche, age at menopause, parity, body mass index, duration of oral contraceptive use and postmenopausal hormones, hypertension, diabetes, and pack–years of tobacco smoking). [Misclassification due to the relative crude definition of night shiftwork was likely to have resulted in bias towards the null.]

2.2.5 Other cancers (Table 2.5)

A cohort of 8603 male full-time manual workers from England and Wales were followed-up during 1956–1968 for cause-specific mortality, including all neoplasms, cancer of lung, stomach, bladder, and leukaemia (Taylor & Pocock 1972). Study subjects were from ten different companies in which they were employed on 1 January 1956. They were born before 1920 and had all been continuously employed at the same company for at least ten years during 1946–1968. Detailed information on all jobs held since 1946 was obtained from company payrolls, including information on working hours, and types of shifts. Based on this information, each worker was allocated into one of three groups: day worker (n = 3860), shiftworker (n = 4188), and ex-shiftworker (n = 555). The criteria for being classified as either a day worker or a shiftworker were that the worker had completed at least 10 years of either work since 1946, with a maximum of 6 months interruption during that period. The term shiftwork covered six types of working hours other than regular day work. The start of follow-up was initiated as soon as each worker met the 10 years of duration of employment criterion. At the end of follow-up, on 31 December 1968, it was possible to trace all but 22 men (0.25%). Information on date and cause of death for the 1578 men who died during the follow-up period was obtained from death certificates. Expected numbers of cause-specific deaths were calculated from the mortality experience of men in England and Wales in 5 year age-groups and calendar time groups. Observed versus expected numbers for all-cause mortality were not significantly different in any of three groups of day workers, shift- and ex-shiftworkers (736/756.4; 722/711; 120/100.9). For the all-neoplasms group, the shiftworkers experienced a significantly higher than expected all-cancer mortality than the general

SHIFTWORK 615

population [SMR, 1.16; 95% CI: 1.02–1.32; 219 observed cases]. For the small group of ex-shiftworkers, 29 deaths were observed versus 25.9 expected [SMR, 1.12; 95% CI: 0.75–1.61], and in the group of day workers, 201 deaths versus 197.1 were observed during follow-up [SMR, 1.02; 95% CI: 0.88–1.17]. For death from cancers of the lung, stomach, bladder, and leukaemia, observed versus expected numbers were, respectively, 94/84.4, 36/25.2, 7/6.6, and 2/3.7 among shiftworkers. Similar patterns were seen in day workers and in ex-shiftworkers. [This study was based on a survivor population with 10 years or more experience of shiftwork which may have underestimated a true increased risk if less than 10 years of shiftwork increased the mortality.]

A census-based ecological cohort study from Sweden included all members of the Swedish population (≥ 20 hours/week) in both 1960 and 1970 (Schwartzbaum et al., 2007). The censuses included individual information about social status and industry but not about work schedules. Therefore, a job–exposure matrix was established for assessment of shiftwork. It relied on a sample of the Swedish population (n = 46 438) collected during 1977–1981, which included information on usual occupation and work schedules. Shiftwork was defined as a schedule with three or more possible shifts per day or work hours during the night for at least one day during the week preceding the interview. About 3% of the men and less than 0.3% of the women participating in the censuses were classified as having done shiftwork, defined by working in industries in 1960 and 1970 where at least 40% of the participants from the survey had reported such a work schedule. Follow-up for cancer in the Swedish Cancer registry was from 1971–1989, and SIRs were calculated on the basis of person–years of follow-up and national rates of specific cancers taken from the Swedish Cancer registry. The SIRs for cancer among men were all close to unity during the 19 years of follow-up, except for kidney (1.14; 95% CI: 1.00–1.31), skin (1.20; 95% CI: 1.02–1.41), and other and unspecified cancers (1.27; 95% CI: 1.07–1.50). For the subgroup of men participating in the 1970 census only, the SIR for thyroid cancer was elevated (1.35; 95% CI: 1.02–1.79). Results changed minimally when the shiftwork status was based only on the 1970 census or other attempts to change the exposure definition. [The major limitation of this study was an unavoidable potential for misclassification of exposure resulting in null results, and to some extent, uncontrolled confounding. Cohort members were followed-up to 1989, although follow-up through 2006 had been possible].

2.3 Aircraft crew Cancer risk of airline personnel has been studied since the 1990s in about ten

countries. The main reason to study these cohorts has been exposure to cosmic radiation, and sometimes passive smoking or electromagnetic fields. Shiftwork as causal factor has not been explicitly mentioned, but in the latest studies, there has be discussion on the potential role of frequent disruptions of circadian rhythm. An alteration in melatonin metabolism decreasing the oncostatic function of this hormone has been hypothesized to be a potential biological mechanism. [It has been questioned whether flight attendants should be considered as shiftworkers.]


For most cabin crew, annual exposure to radiation ranges from 1–6 mSv, compared with approximately 2.4 mSv annually from background radiation. Cosmic radiation includes a substantial neutron component (25–50% of effective dose but less than 5% of absorbed dose). Because flight personnel are the only source of human data on the health effects of exposure to neutron radiation, it is hard to estimate how a large excess risk would be expected due to cosmic radiation. This further makes it difficult to judge how much of the observed excess could be for other risk factors such as shiftwork.

The number of flights over several time zones is used as a proxy of frequency of circadian rhythm disruptions. This number correlates with the dose of cosmic radiation, and therefore estimates of cancer risk in cumulative dose categories can also be interpreted to roughly reflect frequency of circadian rhythm disruptions. On the other hand, separation of the independent roles of these two factors is possible only in large studies with precise information on flight histories. Only one study, combining information on all pilots from the five Nordic cancer registries has been able to make this distinction to a certain extent. In general, the detailed flight histories of airline pilots are known quite well; while for cabin crew, normally only the beginning and end of employment is known. In the airline companies where the principle has been that all cabin crew members fly all routes, an approximation of the radiation dose and numbers of long flights over time zones for each person can be made based on his/her own annual numbers of flight hours, and the flight profile of the company.

All studies published on aircraft crew have been included in this evaluation, irrespective of whether they mention shiftwork or not. Only observations related to breast cancer and prostate cancer have been included in this review, because they are the only ones which have been considered to be associated with shiftwork. The observations related to breast cancer come from cabin crew personnel and those related to prostate cancer mainly from cockpit personnel, because almost all airline pilots are male, and the majority of the cabin crew, female.

In addition to the breast and prostate cancer findings presented below in detail, there is a consistent pattern of increased incidence of skin melanoma and basal cell carcinoma of the skin that are likely to be related to the more frequent sunbathing and sunburns among flight personnel in previous decades. Male cabin crew have also been shown to have a significantly increased risk of Kaposi sarcoma in most studies that included this cancer category. The risk of leukaemia, one of the main target sites in studies on effects of radiation, has been shown to be non-elevated in most studies.

2.3.1 Breast cancer (Table 2.7)

(a) Cohort studies Pukkala et al. (1995) collected a cohort of 1577 all-female flight attendants who had

ever worked for Finnish airline companies (first employment starting in the 1930s). This cohort was followed-up for cancer incidence during 1967–1992. The SIR for breast cancer was 1.87 (95% CI: 1.15–2.23, 20 cases), and the SIR was highest 15–19 years after

SH

IFTWO

RK

617

Table 2.7. Cohort studies of flight personnel and breast cancer


Cohort description Exposure assessment Organ site (ICD code)

Exposure categories No. of cases/ deaths

Relative risk (95% CI)*

Adjustement for potential confounders

Comments

Pukkala et al. (1995), Finland

1577 female cabin attendants who worked for Finnish airline companies; from files of Finnair Flight Company; follow-up for cancer incidence from date of recruitment as cabin crew worker (or January 1967 if later) to emigration, death, or December 1992

Calendar period, length of employment

Breast Any Employment ≥2 years

20 NG

SIR 1.87 (1.15–2.23) 2.0 (1.2–3.2)

Age Control for parity on group level (cohort vs. reference population): parity cannot explain the difference

Lynge (1996), Denmark

915 female airline cabin attendants in Denmark, follow-up for cancer incidence from 1970–1996

Cross-sectional census occupation 1970

Breast Any

14

SIR 1.61 (0.90–2.70)

Age

Wartenberg & Stapleton (1998), USA

287 retired flight attendants from one US airline; follow-up for cancer incidence

Breast Any

7

SIR 2.0 (1.0–4.3)

Age

618 IA

RC

MO

NO

GR

APH

S VO

LUM

E 98

Table 2.7 (contd)






Comments

Haldorsen et al. (2001), Norway

3105 female and 588 male airline cabin attendants with licences issued 1950–1994; follow-up for cancer incidence from 1953 to 1996 (≤38 years)

Total length of employment; length of employment before 26 years of age

Breast Any Employment ≥15 years

38 5

SIR 1.1 (0.8–1.5) 1.0 (0.3–3.0)

Age, number children, age at first birth

Rafnsson et al. (2001), Iceland

1532 cabin attendants, from Icelandic Cabin Crew Association and two airline companies; follow-up for cancer incidence 1955–1997

Year of employment, hired before or in/after 1971

Breast Any

26

SIR 1.50 (1.00–2.10)


Blettner et al. (2002), Germany

16 014 female cabin attendants who had been employed by two German airlines in 1953 or later; mortality follow-up through 1997

Breast Any

19

SMR 1.28 (0.72–2.20)

Age

SH

IFTWO

RK

619

Table 2.7 (contd)






Comments

Reynolds et al. (2002), California, USA

44 021 members of the Association of Flight Attendants in California; follow-up for cancer incidence 1988–1995

Route – international/domestic, length of service, age at entry

Breast Any International Employment ≥15 years Starting age <25 years

60 31 49 41

SIR 1.42 (1.09–1.83) 1.79 (1.21–2.54) 1.57 (1.16–2.08) 1.72 (1.23–2.34)

Age

Linnersjö et al. (2003), Sweden

2324 women from Swedish Scandinavian Airline System employed 1957–1994; follow-up 1961–1996

High altitude, long distance flight hours

Breast Any >5000 block hours in high altitude, long distance flights

33 5

SIR 1.30 (0.85–1.74) Odds ratio 3.27 (0.54–19.7)


Zeeb et al. (2003), European countries

Cabin crew working 1961–1997 in Denmark, Finland, Germany, Greece, Iceland, Italy, Norway and Sweden, a total of 33 063 females and 11 079 male; follow-up for mortality up to 1997

Breast Ever >0–<10 10–<20 years ≥20 years

174 33 19 7

SMR 1.11 (0.82–1.48) 1.12 (0.75- 1.63)1.27 (0.74–2.07 0.80 (0.32–1.77)

Age Overlap with the national incidence studies; breast cancer in women

NG, not given


first employment (SIR, 3.4, 95% CI: 1.5–6.8), and slightly increased with increasing duration of employment. [The flight attendants were more likely than the general population to have multiple reproductive risk factors for breast cancer but these differences were insufficient to explain the magnitude of excesses observed.]

This observation was followed by other observations published in letters in the same journal. Lynge (1996) reported SIRs obtained from a routine tabulation of occupation-specific cancer risks in Denmark. In a 17-year follow-up, the SIR of breast cancer among the 915 women who had reported their occupation in 1970 as a flight attendant was 1.61 (95% CI: 0.90–2.70) when compared to average Danish women. Wartenberg and Stapleton (1998) also reported an increased breast cancer incidence in a small cohort of retired flight attendants from a US airline (SIR, 2.0; 95% CI: 1.0–4.3). This risk seems not to depend on the number of flights, and they suggest that exposure to dicophane (DDT), an organochlorine pesticide used to rid airplanes of insects during 1950–1970, may be a risk factor for breast cancer.

In 2001, two other Nordic studies were published with a setting similar to the earlier Finnish study. Haldorsen et al. (2001) studied a Norwegian cohort of 3144 female flight attendants and observed 38 cases of breast cancer. The SIR was 1.1 (95% CI: 0.8–1.5), and did not increase with increasing length of employment. The authors also had access to the dates of births of the children of the flight attendants for every woman born since 1934. The risk remained largely unchanged after controlling for age at first birth and parity.

Rafnsson et al. (2001) published a cohort study of 1532 cabin attendants from the Icelandic Cabin Crew Association and two airline companies and followed-up for cancer incidence during 1955–1997. The risk of breast cancer was significantly increased (SIR, 1.6, 95% CI: 1.0–2.1, lagged 15 years; SIR, 1.5, 95% CI: 1.0–2.1). Those hired in 1971 or later had the heaviest exposure to cosmic radiation at a young age, and had a signicantly increased risk of breast cancer (SIR, 4.1; 95% CI: 1.7–8.5). The information on reproductive factors among the cabin attendants and the Icelandic female population, obtained from the register of the Genetical Committee of the University of Iceland, provided an opportunity to evaluate the possible confounding due to reproductive factors on the risk of breast cancer in the present study in a similar way as has been recommended when evaluating confounding due to cigarette smoking in occupational studies. Predictive values were calculated on the basis of reproductive factors among the cabin attendants and the population. The risk of breast cancer was 1.0 for parous versus nulliparous, 1.0 for the number of children, and 1.1 for the age at birth of first child.

Rafnsson et al. (2003) also published a case–control study nested in the cohort of Icelandic cabin crew personnel. An increased risk of breast cancer was related to length of employment before 1971, the period before jet aircrafts were taken into operation. The authors concluded that the exposure related to the increased risk of breast cancer was solely confined to the period before 1971, because a long lag time may be required for inducing breast cancer, see Table 2.8. [Their result was compatible with the view that corresponds to a long induction period between ionizing radiation exposure and development of breast cancer.]

SH

IFTWO

RK

621

Table 2.8. Nested case–control studies of airline crew and cancer

Reference, study location and period


Characteristics of cases

Characteristics of controls

Exposure assessment

Exposure categories


Odds ratio (95% CI)


Comments


Breast 35 histologically confirmed cases, including 4 in situ cancers

140 age-matched flight attendants

≥5 versus <5 years work during: pre-jet era (<1971) jet era (≥1971)

29 43

5.24 (1.58–17.4) 0.82 (0.34–1.97)

Age at first delivery, parity

Kojo et al. (2005), Finland

Breast 27 breast cancer cases diagnosed in 1975–2000 among cabin attendants (response proportion of 60%)

517 non-case cabin attendants (response proportion of 52%)

Self-administered questionnaire

Cumulative dose (per 10 mSv). Disruption of menstrual cycle sometimes or often Disruption of sleep rhythm sometimes or often

NG NG NG

0.93 (0.68–1.27) 0.56 (0.12–2.61) 1.52 (0.49–4.74)

Cumulative radiation dose, number of fertile years, parity, family history of breast cancer, alcohol consumption

To assess possible selection bias OR was also calculated for all the subjects in the cabin attendant cohort (44 breast cancer cases and 921 non-cases)

NG, not given


Blettner et al. (2002) studied cancer mortality among 16 014 female and 4537 male cabin attendants who had been employed by two German airlines in 1953 or later. The SMR for breast cancer was 1.28 (95% CI: 0.72–2.20; 19 observed deaths). The SMR did not increase with duration of employment.

In the largest of the studies, Reynolds et al. (2002) linked a group of 44 021 female members of the Association of Flight Attendants in California with the California Cancer Registry. They had to use a computer programme to conduct automated probabilistic record linkage, and to make several assumptions for estimations in the person–year accumulation which may have caused some inaccuracy in the results. During the follow-up, 60 cases of invasive and 12 cases of in-situ breast cancer were recorded during 1988–1995. The SIR for invasive breast cancer across all ethnicities was 1.42 (95% CI: 1.09–1.83) while the incidence of in-situ tumours did not differ significantly from what may have been expected compared to rates from the non-Hispanic Caucasian population or from the population of all races. Invasive breast cancer appeared to be significantly elevated (SIR, 1.79; 95% CI: 1.21–2.54) in flight attendants who were assigned to international routes compared with the general Californian reference population rates.

Linnersjö et al. (2003) observed 33 cases of breast cancer during 1961–1996 among the 2324 women employed at the Swedish Scandinavian Airline System. The SIR of breast cancer was 1.30 (95% CI: 0.85–1.74) when compared to the general population, and did not increase with increasing duration of employment. A case–control analysis nested within the cohort gave ORs for cancer incidence. For cabin crew with at least 5000 block hours in high altitude or long-distance flight hours compared with cabin attendants without this experience, the OR was 3.27 (95% CI: 0.54 −19.7).

Zeeb et al. (2003) reported the combined results from cabin crew cohorts from eight European countries employed during 1921–1997 in Denmark, Finland, Germany, Greece, Iceland, Italy, Norway, and Sweden. During follow-up of 485 831 person–years for women (until 1997), 174 breast cancer deaths were reported (SMR, 1.11; 95% CI: 0.82–1.48).

Kojo et al. (2005) reported on a nested case–control study of breast cancer among Finnish cabin crew attendants. The adjusted ORs were 0.93 (95% CI: 0.68–1.27) for cumulative dose per 10mSV, 0.56 (95% CI: 0.12–2.61) for disruption of menstrual cycle (sometimes or often), and 1.52 (95% CI: 0.49–4.74) for disruption of sleep rhythm (sometimes or often), see Table 2.8.

2.3.2 Prostate cancer (Table 2.9)

Band et al. (1990, 1996) reported cancer incidence and mortality in similar studies among pilots of two Canadian airline companies. Both of them demonstrated an increased incidence from prostate cancer (SIR, 1.54; SMR, 1.87), but only one study demonstrated slightly elevated mortality from prostate cancer (SMR, 1.52; 90% CI: 0.71–2.85; seven deaths) with no dependence on duration of the employment (Band et al., 1996). The authors concluded that detection bias could be a likely explanation, as throughout their

SH

IFTWO

RK

623

Table 2.9. Cohort studies of flight personnel and prostate cancer


Cohort description Exposure assessment




Comments

Band et al. (1990), Canada

913 male pilots employed for 1 year or more by Canadian Pacific Airlines since 1950; cause of death and cancer incidence information up to October 31, 1988 ascertained through the divisions of vital statistics and the cancer registries of the Canadian provinces

Any

6

SIR 1.54 [0.56–3.35]

Age

Band et al. (1996), Canada

2680 male pilots of Air Canada working 1year or more since 1950; cancer incidence and mortality up to 1992.

Any

34

SIR 1.87 [1.30–262]

Age

624 IA

RC

MO

NO

GR

APH

S VO

LUM

E 98

Table 2.9 (contd)






Comments

Irvine & Davies (1992)

446 deaths among serving and retired British Airways (BA) pilots 1966–1989; deaths were ascertained from pension and registry listings, death registries, newspaper obituary listings

BA Personnel and pension records

Pilots

10 10

PMR 2.12 (1.02–3.89) PCMR 1.54 (0.74–2.83)

Age

Irvine & Davies (1999), England & Wales UK

6209 male pilots and 1153 male flight engineers employed since 1939 and for at least 1 year 1950–1992; followed for mortality

Years as flight deck crew, long/shorthaul

Pilots Flight Engineers Longhaul vs shorthaul

15 3 4

SMR 1.11 (0.62–1.84) 0.92 (0.19–2.69) RR 2.47 (0.83–7.65)

Age

Gundestrup & Storm (1999), Denmark

3790 male and 87 female pilots from 1921 and up, followed for cancer incidence 1943–1995

Flight hours Jet Non-jet

3 3

SIR 0.8 (0.2–2.2) 0.8 (0.2–2.2)

SH

IFTWO

RK

625

Table 2.9 (contd)






Comments

Haldorsen et al. (2000), Norway

3815 authorized male pilots 1946–1994; follow-up for cancer incidence up to 1996

Block hours, estimated dose

Any ≥10 000 block hours ≥20 mSv

25 14 6

SIR 1.0 (0.7–1.5) 1.1 (0.6–1.9) 1.8 (0.7–4.0)

Smoking among current pilots compared with reference population (slightly lower proportion among pilots)


458 male pilots 1937–1985; follow-up for cancer incidence 1955–1997

Block hours, estimated dose, crossing time zones (yes/no)

Any International flights

5 4

SIR 1.28 (0.41–2.98) 1.41 (0.38–3.61)

626 IA

RC

MO

NO

GR

APH

S VO

LUM

E 98

Table 2.9 (contd)






Comments

Hammar et al. (2002), Sweden

1490 aircraft pilots and 2808 military pilots and navigators in the Swedish Air Force employed during 1957–1994; follow-up for cancer incidence 1961–l996

Numbers of block hours, high-altitude flights and long-distance flights

Civil Military

18 49

SIR 1.24 (0.74–1.97) 1.17 (0.84–1.49)

Pukkala et al. (2002), Denmark, Finland, Iceland, Norway, Sweden

10 032 male airline pilots employed 1921–1996; follow-up for cancer incidence through 1997

Employment duration, cumulative block hours (by aircraft type), estimated dose

Any >10 000 hours long-haul (as compared with <5000 hours), ages 60+

64 8

SIR 1.21 (0.93–1.54) RR 3.88 (1.26–11.9)

The relative risk of prostate cancer increased with increasing number of flight hours in long distance aircraft (P for trend, 0.01)

SH

IFTWO

RK

627

Table 2.9 (contd)






Comments

Zeeb et al. (2002), Germany

6061 male pilots 1960–1997; mortality follow-up through 1997

Employment durations, cumulative block hours, estimated dose

Any

8

SMR 1.26 (0.5–2.59)

Age

Zeeb et al. (2003), European countries

Cabin crew working 1961–1997 in Denmark, Finland, Germany, Greece, Iceland, Italy, Norway and Sweden, a total of 33 063 females and 11 079 male; follow-up for mortality up to 1997

5 1.09 (0.35–2.68)

628 IA

RC

MO

NO

GR

APH

S VO

LUM

E 98

Table 2.9 (contd)






Comments

Blettner et al. (2003), European countries

Cockpit crew working 1921–1997 in Denmark, Finland, Germany, Great Britain, Greece, Iceland, Italy, Norway and Sweden, a total of 27 797 persons; follow-up for mortality

Any

54

SMR 0.94 (0.71–1.26)

Overlap with the national incidence studies

SHIFTWORK 629

career, pilots underwent yearly physical examinations, including a digital rectal examination.

Irvine and Davies (1992, 1999) studied cancer mortality in British Airways pilots using the proportional mortality ratio (PMR) method (1992), and later (1999) based on SMRs. In the PMR study, there were ten deaths due to prostate cancer. The PMR was 2.12 (95% CI: 1.02–3.89) if the reference was all-cause mortality (excluding aircraft accident), and 1.54 (95% CI: 0.74–2.83) if the reference was all-cancer mortality. [It was however evident from other studies that both overall mortality and all-cancer mortality among airline pilots is markedly below the population mortality rates. In the British Airways pilots (Irvine & Davies, 1999), the SMR was 0.61 for all causes and 0.64 for all cancers, and therefore the PMRs in Irvine & Davies (1992) did not indicate excess prostate cancer mortality among pilots compared to the average population]. In the SMR study (Irvine & Davies, 1999), there were 15 prostate cancer deaths among British Airways pilots (SMR, 1.11; 95% CI: 0.62–1.84), and three deaths among flight engineers often travelling in cockpit (SMR, 0.92; 95% CI: 0.19–2.69). In the internal analysis, the age-adjusted RR between persons flying long-haul versus mainly short-haul (European) flights was 2.47 (95% CI: 0.83–7.65). Flight engineers were assumed to operate in long-haul operations.

Gundestrup & Storm (1999) studied cancer incidence among 3790 male and 87 female commercial Danish cockpit crew members, with records starting from 1921. They were followed for cancer mortality during 1943–1995. Three prostate cancer deaths were observed among both jet pilots and non-jet pilots versus 3.5–4.0 expected.

Haldorsen et al. (2000) published SIRs from a Norwegian cohort of 3815 authorised male pilots employed during 1946–1994. During the follow-up from 1953–1996, 25 cases of prostate cancer were observed (SIR, 1.0; 95% CI: 0.7–1.5); six were in the category of exposed to ≥ 20 mSv (SIR, 1.8; 95% CI: 0.7–4.0).

Rafnsson et al. (2000) studied a cohort of 458 Icelandic male pilots employed during 1937–1985, followed-up for cancer incidence from 1955–1997. There were only four cases of prostate cancer among pilots who had flown international flights (SIR, 1.41; 95% CI: 0.38–3.61), and therefore no dose–response analyses were performed for this cancer site.

The study by Hammar et al. (2002) reported cancer incidence both among civil and military pilots in Sweden. The incidence was about 20% above the national level in both categories, and did not vary with increasing number of block hours, high-altitude flights or long-distance flights.

Zeeb et al. (2002) analysed mortality data of 6061 German male pilots who had worked during 1953–1997. A total of eight deaths from prostate cancer were observed (SMR, 1.26; 95% CI: 0.53–2.59).

Zeeb et al. (2003) reported the combined results from cabin crew cohorts from eight European countries employed during 1921–1997 in Denmark, Finland, Germany, Greece, Iceland, Italy, Norway, and Sweden. During follow-up of 170 634 person–years for men (until 1997), five prostate cancer deaths were reported (SMR, 1.09; 95% CI: 0.35–2.68).


Blettner et al. (2003) combined mortality data from cockpit crew cohorts from nine European countries working during 1921–1997 in Denmark, Finland, Germany, Great Britain, Greece, Iceland, Italy, Norway and Sweden. During the approximately 28 000 person–years of follow-up, 54 prostate cancer deaths were reported (SMR, 0.94; 95% CI: 0.71–1.26).

Pukkala et al. (2002) conducted an analysis among cohorts of airline pilots from Denmark, Finland, Iceland, Norway, and Sweden. When compared to the national studies described above, the length of follow-up was extended. Unpublished Finnish data were added to this analysis. There were 10 032 male airline pilots employed during 1921–1996 who were followed-up for cancer incidence up until 1997. A total of 64 cases of prostate cancer were reported (SIR, 1.21; 95% CI: 0.93–1.54). The RR of prostate cancer increased with increasing number of flight hours on long-haul flights (P for trend = 0.01): the RR was 3.88 (95% CI: 1.26–11.9) for a duration of 10 000 hours long-haul compared with a duration < 5000 hours.

2.3.3 Meta-analyses

Ballard et al. (2000) published a meta-analysis of cancer incidence and mortality among flight personnel. The combined relative risk (meta-RR) based on a fixed effect model for breast cancer was 1.89 (95% CI: 1.40–2.56) overall, or 1.35 (95% CI: 1.00–1.85) if corrected for socio-economic status (based on two studies). The respective meta-RRs for prostate cancer incidence were 1.82 (95% CI: 1.31–2.52) overall and 1.65 (95% CI: 1.19–2.29) if corrected for socio-economic status, also based on two studies.

Seven observational studies among female air cabin crew (Pukkala et al., 1995; Rafnsson et al., 2001; Reynolds et al., 2002; Wartenberg & Stapleton, 1998; Haldorsen et al., 2001; Lynge, 1996) were considered in the review by Megdal et al. (2005) (studies are listed in Table 2.7). With only one exception (Haldorsen et al. 2001), these studies uniformly indicated an increased risk of breast cancer (meta-SIR, 1.44; 95% CI: 1.26–1.65; fixed effects model). All seven studies are incidence studies with the general population as the referent group. [The original rationale for these studies had been that the occupational exposure to cosmic radiation caused an anticipated excess cancer risk. In most studies, the excess starts about 10 years after first employment and increases weakly with increasing duration. Differences in reproductive factors can explain only a small fraction of the excess, and the risk attributable to (non-neutron) radiation, a similar fraction of the excess. Hence, it is likely that there is a major part of the excess risk in breast that must be attributable to factors others than reproductive factors and radiation]. It was reasoned subsequently that the observed increase in breast cancer risk could have been due as well to a melatonin deficiency resulting from work-associated light exposure at night (Mawson, 1998).

Buja et al. (2006) published a similar meta-analysis on cancer risk among female flight attendants, based on the same seven studies. They obtained a meta-SIR of 1.40 for

SHIFTWORK 631

breast carcinoma (95% posterior interval, 1.19–1.65). The only other significant excess was in the incidence of melanoma (meta-SIR, 2.15; 95% CI: 1.56–2.88).

No publication bias was detected by any of the tests used in any of these two meta-analyses.

2.4 References Ballard T, Lagorio S, De Angelis G, Verdecchia A (2000). Cancer incidence and mortality among

flight personnel: a meta-analysis. Aviat Space Environ Med, 71:216–224. PMID:10716165 Band PR, Le ND, Fang R et al. (1996). Cohort study of Air Canada pilots: mortality, cancer

incidence, and leukemia risk. Am J Epidemiol, 143:137–143. PMID:8546114 Band PR, Spinelli JJ, Ng VT et al. (1990). Mortality and cancer incidence in a cohort of commercial

airline pilots. Aviat Space Environ Med, 61:299–302. PMID:2339962 Blettner M, Zeeb H, Auvinen A et al. (2003). Mortality from cancer and other causes among male

airline cockpit crew in Europe. Int J Cancer, 106:946–952.doi:10.1002/ijc.11328 PMID:12918075

Blettner M, Zeeb H, Langner I et al. (2002). Mortality from cancer and other causes among airline cabin attendants in Germany, 1960–1997. Am J Epidemiol, 156:556–565.doi:10.1093/aje/kwf083 PMID:12226003

Buja A, Mastrangelo G, Perissinotto E et al. (2006). Cancer incidence among female flight attendants: a meta-analysis of published data. J Womens Health (Larchmt), 15:98–105.doi:10.1089/jwh.2006.15.98 PMID:16417424

Conlon M, Lightfoot N, Kreiger N (2007). Rotating shift work and risk of prostate cancer. Epidemiology, 18:182–183.doi:10.1097/01.ede.0000249519.33978.31 PMID:17179764

Davis S, Mirick DK, Stevens RG (2001). Night shift work, light at night, and risk of breast cancer. J Natl Cancer Inst, 93:1557–1562.doi:10.1093/jnci/93.20.1557 PMID:11604479

Gundestrup M, Storm HH (1999). Radiation-induced acute myeloid leukaemia and other cancers in commercial jet cockpit crew: a population-based cohort study. Lancet, 354:2029–2031.doi:10.1016/S0140-6736(99)05093-X PMID:10636367

Haldorsen T, Reitan JB, Tveten U (2000). Cancer incidence among Norwegian airline pilots. Scand J Work Environ Health, 26:106–111. PMID:10817375

Haldorsen T, Reitan JB, Tveten U (2001). Cancer incidence among Norwegian airline cabin attendants. Int J Epidemiol, 30:825–830.doi:10.1093/ije/30.4.825 PMID:11511611

Hammar N, Linnersjö A, Alfredsson L et al. (2002). Cancer incidence in airline and military pilots in Sweden 1961–1996. Aviat Space Environ Med, 73:2–7. PMID:11817615

Hansen J (2001a). Increased breast cancer risk among women who work predominantly at night. Epidemiology, 12:74–77.doi:10.1097/00001648-200101000-00013 PMID:11138824

Hansen J (2001b). Breast Cancer Among Women Who Work at Night. Epidemiology, 12:588–599. PMID:11505182

Irvine D, Davies DM (1992). The mortality of British Airways pilots, 1966–1989: a proportional mortality study. Aviat Space Environ Med, 63:276–279. PMID:1610337

Irvine D, Davies DM (1999). British Airways flightdeck mortality study, 1950–1992. Aviat Space Environ Med, 70:548–555. PMID:10373044

Kojo K, Pukkala E, Auvinen A (2005). Breast cancer risk among Finnish cabin attendants: a nested case-control study. Occup Environ Med, 62:488–493.doi:10.1136/oem.2004.014738 PMID:15961626


Kubo T, Ozasa K, Mikami K et al. (2006). Prospective cohort study of the risk of prostate cancer among rotating-shift workers: findings from the Japan collaborative cohort study. Am J Epidemiol, 164:549–555.doi:10.1093/aje/kwj232 PMID:16829554

Lie JA, Roessink J, Kjaerheim K (2006). Breast cancer and night work among Norwegian nurses. Cancer Causes Control, 17:39–44.doi:10.1007/s10552-005-3639-2 PMID:16411051

Linnersjö A, Hammar N, Dammström BG et al. (2003). Cancer incidence in airline cabin crew: experience from Sweden. Occup Environ Med, 60:810–814.doi:10.1136/oem.60.11.810 PMID:14573710

Lynge E (1996). Risk of breast cancer is also increased among Danish female airline cabin attendants. BMJ, 312:253. PMID:8563615

Mawson AR (1998). Breast cancer in female flight attendants. Lancet, 352:626.doi:10.1016/S0140-6736(05)79582-9 PMID:9746034

Megdal SP, Kroenke CH, Laden F et al. (2005). Night work and breast cancer risk: a systematic review and meta-analysis. Eur J Cancer, 41:2023–2032.doi:10.1016/j.ejca.2005.05.010 PMID:16084719

O’Leary ES, Schoenfeld ER, Stevens RG et al.; Electromagnetic Fields and Breast Cancer on Long Island Study Group (2006). Shift work, light at night, and breast cancer on Long Island, New York. Am J Epidemiol, 164:358–366.doi:10.1093/aje/kwj211 PMID:16777931

Pukkala E, Aspholm R, Auvinen A et al. (2002). Incidence of cancer among Nordic airline pilots over five decades: occupational cohort study. BMJ, 325:567.doi:10.1136/bmj.325.7364.567 PMID:12228131

Pukkala E, Auvinen A, Wahlberg G (1995). Incidence of cancer among Finnish airline cabin attendants, 1967–92. BMJ, 311:649–652 http://www.bmj.com/cgi/content/full/311/7006/649. PMID:7549630

Rafnsson V, Hrafnkelsson J, Tulinius H (2000). Incidence of cancer among commercial airline pilots. Occup Environ Med, 57:175–179.doi:10.1136/oem.57.3.175 PMID:10810099

Rafnsson V, Sulem P, Tulinius H, Hrafnkelsson J (2003). Breast cancer risk in airline cabin attendants: a nested case-control study in Iceland. Occup Environ Med, 60:807–809.doi:10.1136/oem.60.11.807 PMID:14573709

Rafnsson V, Tulinius H, Jónasson JG, Hrafnkelsson J (2001). Risk of breast cancer in female flight attendants: a population-based study (Iceland). Cancer Causes Control, 12:95–101.doi:10.1023/A:1008983416836 PMID:11246849

Reynolds P, Cone J, Layefsky M et al. (2002). Cancer incidence in California flight attendants (United States). Cancer Causes Control, 13:317–324.doi:10.1023/A:1015284014563 PMID:12074501

Schernhammer ES, Hankinson SE (2005). Urinary melatonin levels and breast cancer risk. J Natl Cancer Inst, 97:1084–1087.doi:10.1093/jnci/dji190 PMID:16030307

Schernhammer ES, Kroenke CH, Laden F, Hankinson SE (2006). Night work and risk of breast cancer. Epidemiology, 17:108–111.doi:10.1097/01.ede.0000190539.03500.c1 PMID:16357603

Schernhammer ES, Laden F, Speizer FE et al. (2001). Rotating night shifts and risk of breast cancer in women participating in the nurses’ health study. J Natl Cancer Inst, 93:1563–1568.doi:10.1093/jnci/93.20.1563 PMID:11604480

Schernhammer ES, Laden F, Speizer FE et al. (2003). Night-shift work and risk of colorectal cancer in the nurses’ health study. J Natl Cancer Inst, 95:825–828.doi:10.1093/jnci/95.11.825 PMID:12783938

SHIFTWORK 633

Schwartzbaum J, Ahlbom A, Feychting M (2007). Cohort study of cancer risk among male and female shift workers. Scand J Work Environ Health, 33:336–343. PMID:17973059

Skene DJ, Bojkowski CJ, Currie JE et al. (1990). 6-sulphatoxymelatonin production in breast cancer patients. J Pineal Res, 8:269–276.doi:10.1111/j.1600-079X.1990.tb00686.x PMID:2380908

Taylor PJ, Pocock SJ (1972). Mortality of shift and day workers 1956–68. Br J Ind Med, 29:201–207. PMID:5021999

Travis RC, Allen DS, Fentiman IS, Key TJ (2004). Melatonin and breast cancer: a prospective study. J Natl Cancer Inst, 96:475–482.doi:10.1093/jnci/djh077 PMID:15026473

Tynes T, Hannevik M, Andersen A et al. (1996). Incidence of breast cancer in Norwegian female radio and telegraph operators. Cancer Causes Control, 7:197–204.doi:10.1007/BF00051295 PMID:8740732

Viswanathan AN, Hankinson SE, Schernhammer ES (2007). Night shift work and the risk of endometrial cancer. Cancer Res, 67:10618–10622.doi:10.1158/0008-5472.CAN-07-2485 PMID:17975006

Wartenberg D, Stapleton CP (1998). Risk of breast cancer is also increased among retired US female airline cabin attendants. BMJ, 316:1902. PMID:9632420

Zeeb H, Blettner M, Hammer GP, Langner I (2002). Cohort mortality study of German cockpit crew, 1960–1997. Epidemiology, 13:693–699.doi:10.1097/00001648-200211000-00014 PMID:12410011

Zeeb H, Blettner M, Langner I et al. (2003). Mortality from cancer and other causes among airline cabin attendants in Europe: a collaborative cohort study in eight countries. Am J Epidemiol, 158:35–46.doi:10.1093/aje/kwg107 PMID:12835285


3. Studies of Cancer in Experimental Animals

3.1 Modification of carcinogenesis by alteration of light/dark environment and central circadian pacemaker function

Introduction

The regular alternation of light and darkness over 24 hours synchronizes the endogenous circadian timing system of mammals through multiple pathways that involve (a) the elicitation of multiple gene transcription responses to glutamate and pituitary adenylcyclase activating peptide, the neuromediators that convey light messages to the suprachiasmatic nuclei (SCN) in the hypothalamus (Harmar et al., 2002); the SCN are the main circadian pacemakers that coordinate the rhythmic organization of biological functions over 24 hours (Hastings et al., 2003); and (b) the elicitation of multiple hormonal responses, including melatonin and corticosterone, in response to light. These neuroendocrine or endocrine effects seem to require SCN mediation and their extent depends upon the endogenous circadian time of exposure (Ishida et al., 2005; Schibler & Brown, 2005); glucocorticoids, including corticosterone, can reset both the molecular clock and downstream clock-controlled genes in peripheral tissues, such as liver (Balsalobre et al., 2000). In experimental systems, light exposure plays a key role in the resetting of the circadian timing system, that can involve, but does not require, melatonin signalling (Skene et al., 1999).

The relevance of environmental light-dark schedules for cancer development or growth has been studied (see Table 3.1) with regard to the respective effects of (a) circadian time of carcinogen exposure in rodents kept in 24-hour light-dark regimens; (b) constant exposure to light or constant darkness; (c) experimental jet lag or other alterations of photoperiodic regimens; and (d) experimental mutations of clock genes. The intrication between carcinogen effects and circadian disruption was obvious in most reports. Conversely, circadian disruption, through experimental jet lag exposure, as well as clock genes mutations or SCN ablation, also leads to malignant processes taking place.

The influence of circadian time on carcinogen or promoter exposure and malignant growth was studied in intact mice or rats kept in a regular alternation of 12 hours of light and 12 hours of darkness (LD12:12) before exposure to carcinogens. Modifications to the regular alteration of light and darkness over 24 hours can disrupt both the circadian timing system at one or several levels of its hierarchical organization and the clock-controlled rhythms in mammals. The multifaceted changes that can occur range from clear whole organism response such as significant modifications in rest–activity rhythms and inability to rhythmically control core body temperature, to more subtle changes such as altered

SH

IFTWO

RK

635

Table 3.1. Summary of significant positive studies for each type of model and protocol: effect of light exposure during biological darkness and the circadian disruption on cancer incidence and growth

Study type Experimental focus

No other exposure Chemical initiation/ promotion models

Chemical transplacental carcinogenesis models

Tumour cell or graft transplantation studies

Total

Alterations in light exposuresa 2/3b 5/6 1/1 10/10 18/20

SCN lesionsc – – – 1/1 1/1

Chronic experimental jetlag – – – 2/2d 2/2

Pinealectomy-induced melatonin suppression

– 2/8 – 11/13 13/21

Direct effect of physiological concentration of melatonin on tumorigenesis

– – – 5/5 5/5

Clock gene mutations 1/1 1/2e – – 2/3

Circadian timing of carcinogen administration

– 4/4 – – 4/4

Total 3/4 12/20 1/1 29/31 45/56

a Continuous bright light at night, dim light at night, intermittent or pulsed light at night b The one negative study in this category was designed to be negative through the use of an inbred mouse with a genetic predisposition to retinal degeneration and was part of a study with one of the positive findings that had inadequate reporting. c Electrolytic ablation of the superchiasmic nuclei d Both of these studies were performed in the same laboratory with an experimental model that has not yet been used by other groups for cancer studies. e These two studies used a radiation exposure in knockout animals as the cancer-initiating agent rather than a chemical.


activation of rhythmic signal–transduction pathways in multiple cell types in the body (Hastings et al. 2003: Levi & Schibler 2007). The rhythms that exist in the body in such diverse systems such as melatonin release, immune surveillance or cellular proliferation/apoptosis/DNA repair can be altered in different ways by different environmental exposures in different species (Deprés-Brummer et al., 1995; 1997; Fu & Lee, 2003; Filipski et al., 2005). The wide range in effects observed resulting from disruptions of the regular light-dark environment can make interpretation of the reported carcinogenic findings difficult. The rhythm alterations that most often result from chronic changes in the light-dark environment can either persist permanently or be restored through system feedback and adjustment over time. Finally, advancing light onset, as occurs with jet lag, has its own novel impacts on the circadian clock system and the clock-controlled rhythms at all three levels of the circadian timing system: hypothalamic pacemaker, circadian physiology, and molecular clocks (Reddy et al., 2002; Nagano et al., 2003; Filipski et al., 2005). In essence, the circadian disruption that can occur through alterations in the light-dark environment is systematic and can lead to complex phenotypical changes that can only truly be understood in the context of the state of the entire circadian timing system and the downstream pathways it controls.

3.1.1 Chronic alteration of light-dark environment

Groups of 50 2-month old CBA mice, 50% females, were kept under a standard 300 lux light-dark regimen (LD12:12) or a constant 2500 lux light regimen until their natural death. All gross tumours and all tissues and organs with suspected tumour development were examined microscopically. [The Working Group noted that the microscopic examination of all relevant tissues in animals was not done but only done for cases where tumour development was suspected. This may have led to missing microscopic tumours.] No body weight difference between the groups was seen even though there was a significant 30% reduction in food consumption in the constant light group at 6 (P < 0.01), 8 (P < 0.001), 12 (P < 0.02) and 16 (P < 0.02) months. No cataracts were seen in either group. There were no significant changes in length of the estrous cycle, although mice in the constant light group were more likely to have irregular cycles at 3 (P < 0.05), 6 (P < 0.001) and 12 (P < 0.002) months. The total number of animals with malignant tumours was significantly increased in the constant light group (35% versus 10% of tumour-bearing mice; P < 0.001) as was the incidence of lung adenocarcinomas (7/50 (14%) versus 1/50 (2%); P < 0.05) and malignant lymphomas and leukaemia combined (6/50 (12%) versus 0/50; P < 0.02). There was also a marginal increase in hepatocellular carcinomas (4/50 (8%) versus 0/50; P = 0.058) (Anisimov et al., 2004).

SHIFTWORK 637

3.1.2 Role of circadian time on two-stage models of carcinogenesis

(a) Two-stage skin cancer carcinogenesis mouse model

(i) N-Nitroso-N-methylurea Six groups of 77 to 105 hairless mice (hr/hr Oslo strain), 60–90 days of age, were

kept in LD12:12 (06:30 to 18:30) and painted with N-nitroso-N-methylurea (NMU, 0.2 mg) once at 08:00, 12:00, 20:00, and 24:00 or once a week for 3 weeks at 08:00 and 20:00. There were equal numbers of male and female mice in each group. The mice were examined for appearance of skin papillomas once a week for 18 months. No difference was found between animals painted once at 05:30 or at 17:30 hours after light onset. The number of animals having tumours from the remaining four groups were 5/95 (5%) in the group painted once at 01:30, 6/77 (8%) in the group painted three times at 01:30, 10/104 (10%) in the group painted once at 13:30 and 13/96 (14%) in the group painted three times at 13:30. The difference between tumour incidence for the two single painting groups (01:30 and 13:30) were not significantly different (P = 0.105), the difference in the groups painted three times was marginal (P = 0.054). When the 01:30 groups were combined and compared to the combined 13:30 groups, the difference was significant (P = 0.0137). The authors noted that the highest incidence of papillomas was achieved when exposure to NMU occurred at a time corresponding to the lowest DNA synthesis rate when relatively large numbers of late G1 cells accumulated [no data presented] (Clausen et al., 1984).

A total of 670 hairless mice (hr/hr Oslo strain, 50% females) kept in LD12:12 with light from 07:30 to 19:30 were exposed to a single topical application of two doses of NMU dissolved in 100 μL reagent-grade acetone. Mice (n = 351) were exposed in groups to a single application of 1 mg NMU at either 00:00, 04:00, 08:00, 12:00, 16:00 or 20:00. Also, several mice (n = 287) were exposed in three groups to a single application of 2 mg NMU at 08:00, 12:00 or 20:00. An additional group of 32 mice were also treated with 10 mg NMU to study the dose–response relationship. The development of all types of skin tumours was observed and the results presented as tumour rates (percentage of tumour-bearing animals in relation to the number of animals alive, appearance of the first tumour related to time), and tumour yields (the cumulative occurrence of all skin tumours standardized for comparison of groups of 32 mice related to time). Most animals were examined once a week for 54 weeks, but those to which 10 mg NMU was applied were observed for only 34 weeks. A circadian variation in tumour production after a single application of 1 mg NMU was demonstrated with a high tumour incidence after application in the period from 16:00 to 00.00, and a lower incidence between 04:00 and 12:00. When 2 mg NMU was applied, there was definitely a low tumour incidence after application at 04:00 compared to the two other times. There was a good and almost straight-lined dose–response relationship after the application of 1, 2, and 10 mg NMU. [Publication unavailable to the Working Group] (Iversen & Iversen, 1995).


(ii) 7,12-dimethylbenz[α]anthracene/12-O-tetradecanoylphorbol-13-acetate

Groups of 25 female CD1 mice, 8–10 weeks of age, were kept in LD12:12, with light onset at 06:00 and were administered a single dose of 9,10-dimethyl-1,2-benzanthracene (DMBA, 2.5 μg/mouse). Starting 2 weeks later, the back of the mice were painted twice a week for 12 weeks with tumour promoter 12-O-tetradecanoylphorbol-13-acetate (1.9 μg). The application of 12-O-tetradecanoylphorbol-13-acetate was tested at: 05:00, 11:00, 17:00 (during the light span), and 23:00 (5 hours after the onset of darkness). The experiment was terminated at Week 15 following the first 12-O-tetradecanoylphorbol-13-acetate administration. In the group painted at 23:00 as compared to that exposed at 17:00, there were nearly twice as many mice with tumours within Weeks 8–11, and the weekly average number of tumours per mouse was nearly twice as high within Weeks 9–15 (Wille, 2003). [The Working Group noted that the tumour type was never clearly given but was assumed to be skin papilloma or carcinoma.]

(b) Intestinal carcinogenesis models in rats

(i) 1,2-dimethylhydrazine Male non-inbred rats from a local strain kept in LD12:12 were injected

subcutaneously with five weekly doses of 21 mg/kg body weight 1,2-dimethylhydrazine 1 hour after light onset (10:00; n = 23) or 13 hours after light onset (22:00; n = 24). Injections at 13 hours after light onset were followed by a significant decrease in the incidence (from 91 to 75%; P < 0.05) and mean size of tumours (from 27.2 to 15.5 mm2; P < 0.05) 6 months after the first injection; there were also relatively fewer large tumours (Dubina et al. 2002).

(ii) Azoxymethane The effect of the time of administration of azoxymethane during the day to alter the

yield of foci (recognized as precancerous lesions) in the colon was evaluated in six groups of 14 male Fischer 344 rats, 7 weeks of age. Azoxymethane was subcutaneously administered in a 15 mg/kg body weight saline solution on Days 7 and 14 of the experiment. Foci of aberrant crypts were evaluated with microscopy 28 days after the first azoxymethane dose in whole mounts of colons stained with methylene blue. In each of the three groups of rats, azoxymethane induced twice as many foci when administered between 8.40 and 11.00 than in the three other groups that were administered azoxymethane between 14.45 and 17.55 (205.7 ± 16.0 versus 110.2 ± 12.9; ANOVA, P < 0.01). [No indication as to the light-dark schedule in the animal rooms was given by the authors] (Pereira et al., 1994).

Groups of 30 female mice from two inbred substrains of C3H mice, 100 days of age, were exposed to constant light [intensity not given] or LD12:12 (control) over 400 days. In substrain C3H-A, lifetime exposure to chronic light extended estrus by approximately 24 hours, and resulted in a significant increase in the incidence of and mortality from

SHIFTWORK 639

mammary tumours [data only given in the form of a graph with no statistical details; 50% versus 0% and 70% versus 20% mammary tumour incidence at 151–200 and 201–250 days of age, respectively]. In substrain C3H-HeJ [this substrain has an inherited autosomal recessive retinal degeneration and has been well characterized in circadian behavioural paradigms], lifetime exposure to chronic light resulted in permanent estrus [onset time not given], a delayed onset of mammary tumour development, a reduced number of tumours, and an increased longevity [data only given in the form of a graph with no statistical details] (Jöchle, 1963; Joechle 1964).

3.1.3 Constant light, light during darkness and constant darkness on experimental cancer

(a) DMBA on mammary cancers in female rats A group of 28 female non-inbred virgin rats was injected intravenously with 1.5 mg

DMBA in saline emulsion six times once every 10 days. Four weeks after the last injection of the carcinogen, animals were divided into two equal groups. Females of the first group were exposed to constant electric light [300-W electric bulb], while the animals of the second group were kept under conditions of natural light. Animals of the constant light group had an increased multiplicity of mammary gland tumours (3.1 versus 1.7). Altogether, tumours developed in 12/14 rats (86%) in the first group and in 10/14 rats (71%) in the second group. In a second experiment, exposure to constant light that started 7 weeks before DMBA treatment resulted in mammary gland tumours in 5/17 rats (29%) while 6/15 females (40%) animals kept under natural light developed tumours (Khaetski, 1965).

Female Sprague-Dawley rats were exposed to LD12:12 or constant illumination starting on the 42nd day of life. At 50 days, the animals received 30 mg DMBA intragastrically. Eight months later, mammary tumours were found in 15/26 rats (58%) in LD12:12, with 16/36 (44%) tumours being adenocarcinomas. Conversely, 20/21 rats (95%) on constant light displayed mammary tumours, however, 53/57 (93%) tumours were fibroadenomas and only 4/57 (7%) were adenocarcinomas. Constant light reduced (P < 0.001) the weights of both the ovaries and pineal glands (Hamilton, 1969). [No information was given on the time of DMBA administration, sampling times or circadian effects of constant light.]

Pregnant female Holtzman rats, 10–12 days of age, were exposed to LD10:14 or to constant light, with light intensity of 150 lux, and their pups were subsequently exposed to the same regimen. The female offsprings were fed with 100 mg/kg DMBA intragastrically on Days 55 to 60. This procedure led to 20% mortality. Animals were palpated once weekly. Over the 180 days that followed DMBA delivery, 60% of the 25 rats in LD10:14 and 94% of the 47 rats in constant light developed tumours (P < 0.02). Of these tumours, 60% were adenocarcinomas in the LD10:14 group as compared to 95% in the constant light group (P < 0.001). The latency period of tumour appearance was also


significantly shorter in the constant light group (P < 0.001). Constant light exposure was associated with a 3-fold increase in plasma prolactin and a 5-fold increase in DNA synthetic activity in the mammary gland. Minor changes in plasma estradiol were found between the LD10:14 and constant light groups (Shah et al., 1984; Kothari et al., 1982, 1984; Mhatre et al., 1984). [No information was given on the time of DMBA administration, sampling times or circadian effects of constant light, three factors that can help in the interpretation of these clearcut differences in carcinogenesis.]

A total of 100 female Sprague-Dawley rats were divided by weight into two groups of 50 animals. Starting at Day 26, one group was exposed to constant light, and the second group was exposed to LD8:16. Both groups received an 8 mg intragastric dose of DMBA at 52 days of age. At 13 weeks post-DMBA administration, there were significantly fewer mammary tumours in the constant light group compared with the LD8:16 group: the respective number of tumour-bearing rats were 8/50 (16%) vs 19/50 rats (38%) (P < 0.05), with an average mean number of tumours per rat of 1.1 and 2.6, respectively. The protective effect of constant light on tumorigenesis was ascribed to a substantial acceleration of mammary-gland development, past the temporal window of DMBA susceptibility in virgin animals. This was supported by the observation that 29/50 rats (58%) on constant light had lactating mammary nodules at necropsy compared to 0/50 in LD8:16 (Anderson et al., 2000).

The effects of four light–dark schedules on the growth of established mammary carcinomas was investigated using 64 female Sprague Dawley rats, exposed to a single intragastric dose of 20 mg/kg DMBA at 55 days of age. When the number of rats with palpated mammary tumours of size 1-cm in diameter was sufficient, the rats were randomized into four groups of 16 animals each. These groups were subjected to LD12:12, constant light (300 lux), LD12:12 (300 lux) with a 30-minute exposure to light near mid-dark, or exposed to low light intensity (0.21 lux) rather than real darkness during the 12-hour dark span. Tumour growth was fastest in the rats exposed to dim light during the dark phase, and slowest in those remaining in LD12:12 (P < 0.01). Tumours also grew faster (P < 0.05) in the rats exposed to light at night for 30 minutes and in those exposed to constant light, when compared to the LD12:12 group. The onset of accelerated tumour growth occurred at Week 4 for the dim-light-at-night group, at Week 6 for the light-pulsed-at-night group, and at Week 10 for the constant light group. On Weeks 11 and 12, the mean tumour surface of the three experimental groups with altered light–dark schedules were similar and differed significantly from the control LD12:12 group (P < 0.05). A significant decrease in night time urinary excretion of aMT6s was observed in all experimental groups. The usual light/dark difference in aMT6s excretion was eliminated in the dim-light-at-night and constant light groups but not in the light-pulsed-at-night group. A significant 5-fold increase in serum estradiol was observed in the dim-light-at-night group only (Cos et al., 2006).

SHIFTWORK 641

(b) NMU The effects of altered endogenous night time melatonin concentrations on mammary

tumour production were investigated in an NMU-induced breast cancer model in female Fischer 344 (F344)/N rats, as were the effects of suppressed serum melatonin concentrations on the incidence and progression of NMU-induced breast cancer. In-vivo studies were used to assess serum melatonin concentrations and the incidence of NMU-induced tumours after 1 day, 2, and 10 weeks of nightly administration of short-duration intermittent light exposure at night. Five 1-minute exposures to incandescent light every 2 hours after the start of the dark phase of the light–dark cycle decreased the magnitude of the nocturnal rise of serum melatonin concentrations in rats by approximately 65%. After 2 weeks of nightly intermittent light exposures, the peak night time serum melatonin concentrations decreased by approximately 35% on average. The amelioration continued and, at 10 weeks, peak night time serum melatonin concentrations had decreased further, by approximately 25%. Because peak endogenous night time serum melatonin values could be moderately suppressed for at least 10 weeks, a 26-week NMU mammary tumour experiment was conducted. Serum melatonin concentrations and incidence, multiplicity, and weight of NMU-induced mammary tumours were assessed. No effect on the development of mammary tumours in an NMU-induced tumour model in rats occurred when endogenous night time serum melatonin concentrations were moderately suppressed by short-duration intermittent light exposures at night. At necropsy, there were no alterations in mammary tumour incidence (28/40 NMU controls (70%), 28/40 NMU + light (70%)), multiplicity (2.18 tumours/tumour-bearing NMU control, 1.89 NMU + light), or average tumour weight (1.20 g NMU control, 1.19 g NMU + light). Tumour burden had no effect on the serum melatonin cycle. At 26 weeks, however, animals exposed to intermittent light at night exhibited an approximately 3-fold higher serum melatonin concentrations when compared to controls (Travlos et al., 2001).

(c) Diethylnitrosamine on liver carcinogenesis in male rats A total of 65 male Wistar rats were administered diethylnitrosamine (10 mg/kg/day

p.o. in drinking-water) for 6 weeks and were randomized into three groups. Rats received either diethylnitrosamine only (n = 20), phenobarbital (n = 22, 30 mg/rat/day p.o. in drinking-water) for 4 weeks as a promoting agent or were exposed to constant light (n = 23). All three groups received diethylnitrosamine in LD12:12 for the initial 21 days and in constant light from Days 22 to 43. This procedure was chosen to suppress the main circadian physiology outputs expected to occur in the continuous light group by Day 77, simultaneously with the maximum phenobarbital promoting effect expected to occur in the phenobarbital group. In the constant light group, there was a 4-fold drop in 24-hour mean urinary aMT6s excretion when compared to the other groups (P from ANOVA < 0.001). The aMT6s rhythms were suppressed in constant light, but remained similar in the other groups. Laparotomy was then performed and macroscopic nodules were counted and measured in each of the four liver lobes, using a 1-mm scale, without knowing the


allocated group of the animal. The proportion of rats with macroscopic nodules was 72% (LD12:12 group), 89% (phenobarbital group), and 95% (constant light group) (P from χ2 = 0.10). Both the frequency of the lesions and the size of the largest lesion differed significantly among the three groups (P < 0.05). Nodules were more numerous and larger in the constant light group and in the phenobarbital group when compared to the LD12:12 group (P from χ2 < 0.05). Similarly, there were more rats with large tumours (≥ 3 mm) in the constant light group and in the phenobarbital group when compared to the LD12:12 group (P < 0.05). All the rats died with hepatocellular carcinomas, with a median survival of 5 months – this was similar in all three groups. (Van den Heiligenberg et al., 1999). [Light-induced circadian clock suppression exerted a promoting effect similar to that caused by phenobarbital in this model. Reduced body weight through manipulation of feeding schedules has usually been associated with decreased tumour growth. Animals on constant light had increased incidence and volume of tumours and gained the least weight.]

(d) Transplanted Glasgow osteosarcoma in mice Three groups of 10 male B6D2F1 mice were randomly assigned to remain in

LD12:12 or exposed to constant darkness or to constant light throughout the whole study duration (5 weeks). Environmental conditions were confirmed to maintain circadian coordination in mice, with a period of 23.8 hours for the constant darkness group, and 26.5 hours for the constant light group. After 3 weeks, all the mice were inoculated subcutaneously with 3x3-mm fragments of mouse Glasgow osteosarcoma in each flank. Tumour size was measured three times weekly for 2 weeks. Exposure to constant darkness or to constant light had no significant effect on tumour growth (ANOVA: P = 0.8) nor survival (log-rank: P = 0.66) when compared to mice kept in LD12:12 (Filipski et al., 2004). [The Working Group noted that in B6D2F1 mice, constant light and constant darkness did not suppress circadian outputs or melatonin secretion, explaining why the study was negative.]

(e) Tissue-isolated Morris rat hepatoma 7288CTC Young adult male Buffalo rats (BUF/BUF/Ncr), 5 weeks of age, were allowed to

acclimatize to a photoperiod of LD12:12 for 1 week, and were provided with standard laboratory chow and water ad libitum. Rats were randomly assigned to three different light exposure groups (n = 12 rats/group): LD12:12 (1st group), LD12:12 with a dark phase contaminated with dim light coming from an indirect light leak of 0.2 lux (2nd Group) through the door of the final animal group of constant bright light of 810 lux (3rd Group). Half of the animals were exposed to these lighting conditions for 4 weeks after which blood samples were collected at 4-hour intervals over a 24-hour period. The remaining animals were exposed to these lighting conditions for 2 weeks before their subcutaneous implantation of a single 3 mm cube of MT1/MT2 melatonin-receptor positive Morris rat hepatoma 7288CTC in a tissue-isolated manner (n = 6 rats/group).

SHIFTWORK 643

Tumour growth (estimated tumour weights calculated from serial measurements of tumour size) was monitored throughout the course of exposure to these different lighting conditions for up to 3 additional weeks following tumour implantation. At the end of each tumour growth period in each light exposure group, tumour linoleic acid uptake and 13-hydroxyoctadecadienoic acid (13-HODE) release were measured via arteriovenous difference measurements; tumours were removed for analysis of linoleic acid content. Compared with the 1st group in which the nocturnal circadian increase in plasma melatonin and linoleic acid levels were intact, the mean tumour growth rate was increased over 2-fold (P < 0.05) and tumour latency decreased by 6 days [no statistical analysis peformed] in the 3rd group in which the nocturnal rise in the levels of both plasma melatonin and linoleic acid were extinguished. Plasma melatonin levels were consistently low while plasma linoleic acid levels remained the same throughout the 24-hour day in the 3rd group. Tumour linoleic acid uptake was increased 2-fold while 13-HODE release was augmented over 4-fold when compared to 1st group (P < 0.05). In the 2nd group, the mean tumour growth rate was increased nearly 2-fold (P < 0.05) and tumour latency decreased by 2 days [no statistics performed] when compared with the 1st group; the plasma profile of low plasma melatonin levels was similar to the 3rd group while the nocturnal circadian rise in plasma linoleic acid levels was preserved, as per the 1st group. Tumour linoleic acid uptake was increased 2-fold while 13-HODE release was augmented 3-fold when compared to the 1st group (P < 0.05). Under all three lighting conditions the total consumption of food throughout the day was virtually identical (Dauchy et al., 1997).

In a follow-up confirmatory study of essentially identical design as that described above, virtually the same results were obtained on the growth, linoleic acid uptake and 13-HODE release in tissue-isolated rat hepatoma 7288CTC. The only substantial difference between the two studies was that the source of dim light exposure during the dark phase was from direct illumination (0.2 lux) provided by a mounted fluorescent lighting fixture placed directly in front of the animal cages rather than from an indirect, contaminating light leak in the previous study (Dauchy et al., 1997).

In a light intensity dose–response study, adult male Buffalo rats were acclimatized to an LD12:12 photoperiod and then randomly assigned to one of six different light intensity exposures during the dark phase of an LD12:12 photoperiod in specially constructed light exposure chambers: total darkness, 0.02, 0.05, 0.06, 0.08 and 345 μW/cm2; n = 6 rats per exposure. Indirect white, polychromatic fluorescent light reflected off of the walls of the chamber were used rather than direct light from the fluorescent tubes. Rats were exposed to these different lighting conditions for 2 weeks before the implantation of tissue-isolated Morris rat hepatoma (MT1/MT2 melatonin-receptor positive) 7288CTC. Following tumour implantation, rats were maintained on their respective light exposure regimen until the end of their respective tumour growth periods. Following 2 weeks of exposure, nocturnal serum levels of melatonin were suppressed in a dose–response manner until full suppression was reached at the highest light intensity. There was a marked dose-related increase in tumour growth rates, [3H]thymidine incorporation into DNA, and DNA


content relative to the control animals exposed to the dark phase over a period of up to 3 weeks following tumour implantation (P < 0.05). Similary, tumour linoleic acid uptake, 13-HODE release, cyclic adenosine monophosphate (cAMP) levels (P < 0.05), extracellular signal-regulated kinase kinase (MEK) and extracellular signal-related kinase (ERK1/2) activation [no statistical analysis] were markedly increased as the light intensity during the dark phase increased. No significant dose–response effects of light were observed on the serum levels of corticosterone (Blask et al., 2005).

(f) Tissue-isolated MCF-7 human breast cancer xenografts in female nude rats

Adult female inbred nude rats maintained on an LD12:12 photoperiod were implanted with estrogen/progesterone-receptor positive (ER+/PgR+) and MT1/MT2 melatonin-receptor positive MCF-7 human breast cancer xenografts in a tissue-isolated manner. Tumour growth was monitored for 40 days (estimated tumour weight of 2.5 g) at which time a subgroup of tumour-bearing rats (n = 3) was tranferred to constant bright fluorescent white light (300 lux) while the remaining rats (n = 4) were maintained on an LD12:12 photoperiod for the duration of the tumour growth period. Serum melatonin levels were measured in parallel groups of rats during the mid-day and mid-dark phases of an LD12:12 photoperiod, and during the subjective dark phase during constant bright light exposure (n = 6/group) after 5 weeks of being maintained under these conditions. In the group of tumour-bearing rats switched from LD12:12 to constant light, which induced a complete suppression of nocturnal circulating melatonin levels, the tumour growth rate increased (P < 0.05) 7-fold when compared to the control group maintained on LD12:12, and exhibited a robust nocturnal melatonin rise in the blood (P < 0.05). Tumour linoleic acid uptake increased 2-fold whereas 13-HODE production increased over 5-fold in the group exposed to constant bright light relative to the LD12:12 controls (P < 0.05) (Blask et al., 2003).

Identical in design to the above study, a light intensity dose–response study was performed in adult female nude rats, implanted with tissue-isolated ER−/PgR− and MT1 melatonin receptor positive MCF-7 human breast cancer xenografts (Blask et al., 2003). Two weeks of exposure of rats to increasing intensities of white light before tumour implantation resulted in nocturnal serum levels of melatonin that were suppressed in a dose–response manner until full suppression was reached at the highest light intensity. Continued exposure of these tumour-bearing rats to the same conditions of increasing light intensities over a period of up to 5 weeks following tumour implantation resulted in a marked dose-related increase in tumour growth rates, [3H]thymidine incorporation into DNA, and DNA content relative to the control animals exposed to the dark phase (P < 0.05). Similarly, tumour linoleic acid uptake, 13-HODE release and cAMP levels were significantly increased (P < 0.05) whereas extracellular MEK and extracellular ERK1/2 activation were markedly increased [no statistical analysis]. No significant dose–response effects of light were observed on the serum levels of either estradiol or corticosterone (Blask et al., 2005).

SHIFTWORK 645

3.1.4 Effect of experimental chronic jet lag on cancer in the mouse

Two groups of 16 male B6D2F1 mice each were assigned randomly to remain in standard lighting (LD12:12) or to be exposed to experimental chronic jet lag (through serial 8-hour advances of LD12:12 cycles every 2 days). This schedule was considered as being the most disruptive on the nest-activity circadian rhythm among schedules tested in previous experiments. The locomotor activity and body temperature of the mice were monitored with a radiotransmitter. Ten days after the start of light–dark cycle advances, animals in both groups were inoculated subcutaneously with a 3 × 3-mm fragment of transplantable mouse Glasgow osteosarcoma. Three mice in each group served as non-tumour-bearing controls. Survival was checked daily and tumour size measured with a caliper three times a week. Tumour weight was computed as (length × width2)/2. Fifteen days after tumour inoculation, mice were sacrificed following exposure to constant darkness for up to 48 hours at four different circadian times, i.e. 0, 6, 12 or 18 hours after light onset. Tumour and liver samples were taken to measure the circadian expression of the clock genes mPer2 and mReverbα with an RNase protection assay (Filipski et al., 2004). This experiment was replicated using 12 control mice in LD12:12 and 14 animals subjected to experimental chronic jet lag. Tumour size was measured daily or every alternate day. Tumours progressed significantly faster in animals undergoing “jet lag” when compared to those kept in LD12:12 in both experiments (ANOVA: P < 0.001 and P = 0.002, respectively). Both experiments indicated that chronic jet lag accelerated tumour growth predominantly between the 8th and the 11th day following tumour inoculation. In the first experiment, mean tumour weight (± SEM) on Day 11, i.e. before the death of the first animal, was 1330 ± 151 mg in experimental jet lag mice, and 647 ± 56 mg in controls (t-test: P = 0.001). In the replicated experiment, it was 1376 ± 131 mg, and 847 ± 107 mg in experimental jet lag and control mice, respectively (t-test: P = 0.005). The survival curves further differed with statistical significance as a function of lighting schedule with poorest survival seen in the experimental jet lag group, in each experiment considered separately (log-rank: P = 0.013 and P = 0.0025), or pooled (P < 0.0001) (Filipski et al. 2004).

The same protocol was applied to 13 male B6D2F1 mice on LD12:12 and to 14 mice undergoing experimental chronic jet lag for the previous 10 days. On Day 12, mean tumour weight was 1317 mg in the LD12:12 group and 1997 mg in the experimental chronic jet lag group (P = 0.04). Experimental chronic jet lag suppressed the circadian rhythms in mRNA transcription of the clock genes Rev-erbα and Bmal1, while dampening and phase-shifting that in Per2 in the liver of male B6D2F1 mice. This resulted in a significant derepression of c-Myc that became profoundly rhythmic, while P53 was repressed. Both effects favour genomic instability and cellular proliferation [two effects that could favour liver carcinogenesis]. In the tumours of mice kept in LD12:12, the mRNA expression patterns of the clock genes was suppressed for Rev-erbα and markedly damped for Per2 and Bmal1, when compared with liver. Experimental chronic jet lag suppressed these mRNA rhythms in tumours (Filipski et al., 2005).


3.1.5 Endogenous circadian disruption on experimental cancer

(a) SCN ablation in mice The SCN of male B6D2F1 mice were destroyed by bilateral electrolytic lesions, and

body activity and body temperature were recorded with a radiotransmitter implanted into the peritoneal cavity. Mice were inoculated subcutaneously with 3 × 3-mm fragments of mouse Glasgow osteosarcoma tumours (n = 16 with SCN lesions, n = 12 sham-operated) or pancreatic adenocarcinoma tumours (n = 13 with SCN lesions, n = 13 sham-operated) to determine the effects of altered circadian rhythms on tumour progression. Complete SCN destruction was ascertained postmortem. Both types of tumours grew two to three times faster in mice with SCN lesions when compared to sham-operated mice (P < 0.001). The survival of mice with SCN lesions was significantly shorter compared with that of sham-operated mice (log-rank P = 0.0062). The 24-hour rest–activity cycle was ablated and the daily rhythms of serum corticosterone level and lymphocyte count were markedly altered in 75 additional mice with complete SCN destruction when compared to 64 sham-operated mice (P < 0.001). Thus, disruption of circadian rhythms in mice was associated with an accelerated growth of malignant tumours of two types, suggesting that the host circadian clock may play an important role in the endogenous control of tumour progression (Filipski et al., 2002).

(b) Clock gene mutations in mice Fu et al. (2002) observed that knock-out mice without expression of mPer2 (mPer2m/m

mice) displayed salivary gland hyperplasia in both males and females and teratomas, predominantly of the epidermis in males at 6 months, with no other apparent pathological defect. By the age of 12 months, all mPer2m/m mice showed salivary gland hyperplasia, and all male mPer2m/m mice developed teratomas around the genital areas. In addition, 30% of the 34 mPer2m/m mice in the study died before the age of 16 months and 15% of the mPer2m/m mice died of lymphoma, an event that was not observed before the age of 20 months in the 40 wild-type mice studied (P < 0.001).

To examine further the role of mPer2 in suppressing neoplastic growth, wild-type and mPer2m/m mice at 8 weeks of age were challenged with a single dose of whole-body radiation of 4 Gy 10 hours after light onset, and were monitored for illness and survival. The mPer2m/m mice were more sensitive to gamma radiation, as indicated by premature hair graying and hair loss, and an increased rate of tumour formation. Hair graying was observed in 50% of mutant mice at 12 weeks after irradiation, a difference that held up also at 22 weeks after irradiation. The irradiated mPer2m/m mice also showed an earlier onset of hyperplastic growth. At 7 months after irradiation, teratomas were observed in all irradiated male mPer2m/m mice, but not in any irradiated wild-type mice. At 16 months after irradiation, 71% of the mPer2m/m mice had developed malignant lymphomas, with a first case discovered at 5 months. Malignant lymphomas were found in multiple organs – liver, lung, spleen, heart, intestine, salivary glands, etc. Conversely, only 5% of the wt

SHIFTWORK 647

mice displayed malignant lymphomas 16 months after irradiation. [The core circadian genes are induced by gamma radiation in wild-type mice but not in mPer2 mutant mice. Temporal expression of genes involved in cell cycle regulation such as cyclin D1, cyclin A, Mdm-2, and Gadd45α, is deregulated in mPer2 mutant mice. In particular, the transcription of c-myc is controlled directly by circadian regulators and is deregulated in the mPer2 mutant. In this study, c-myc transcription in liver was derepressed and became rhythmic in mPer2m/m mice as compared to wt animals.] In the mPer2m/m mice, P53 was repressed when compared to wt. These studies suggested that the mPer2 gene functions in tumour suppression through regulating DNA-damage-responsive pathways (Fu et al., 2002).

Gauger and Sancar (2005) studied Cry1–/–/Cry2–/– mice and fibroblasts derived from these mice for radiation-induced cancer and killing, as well as DNA-damage checkpoints and killing, respectively. They administered a single dose of 4 Gy to 24 wt C57/Bl6 and to 27 double mutant mice kept in LD12:12, 10 hours after light onset. No difference in survival was found between the two groups over the 90 weeks following radiation exposure and no overt lymphoma or other tumour was found in these mutant mice. Similarly, the Cry1–/–/Cry2–/– mutant fibroblasts were indistinguishable from the wild-type controls with respect to their sensitivity to ionizing radiation and UV radiation and ionizing-radiation-induced DNA damage checkpoint response. According to the authors, their data suggest that disruption of the circadian clock in itself did not compromise mammalian DNA repair and DNA damage checkpoints and did not predispose mice to spontaneous and ionizing-radiation-induced cancers (Gauger & Sancar, 2005).

[A single timepoint of radiation exposure was tested here. No demonstration was offered in this study that Cry1–/–/Cry2–/– mice had an ablated circadian clock. The in-vivo part of this study was performed in mice kept in LD12:12, an environmental condition that dampens, yet does not disrupt, 24-hour physiology in Cry1/Cry2 double mutants, while constant darkness exposure does (Nagashima et al., 2005). The Working Group noted that recent data show that Crys do not seem to be required for normal circadian clock function in mouse fibroblasts (Fan et al., 2007).]

3.1.6 Transplacental carcinogenesis

Three groups of 24 pregnant Wistar rats were exposed to different light–dark regimens consisting of either LD12:12 (control), constant darkness, or constant light from Day 1 of pregnancy. On the 18th to 19th day of pregnancy, the dams were injected with a single intravenous dose of 80 mg/kg N-nitrosoethylurea, a chemical known to cause tumours of the peripheral nervous system and kidney when administered under these conditions. Dams from the continuous darkness group were subjected to approximately 10 minutes of light during injection. All pups were kept with their dams during the lactating period (1 month) where they remained in the dams’ initial light–dark regimen. Following lactation, pups were removed from the dams, housed in groups of 5–7 animals separated by sex and kept under the 12:12 light–dark cycle until their natural death. [The


rats were kept in this regimen since their conception until one month of age.] Full necropsies were performed on all animals with all suspected tumours and all tissues suspected for tumour growth examined microscopically. [The Working Group noted that the microscopic examination of all relevant tissues in animals was not systematically done and only for cases where tumour development was suspected. This may have led to mining microscopic tumours.] The incidence of any tumour, tumours of the peripheral nervous system, and tumours of the kidney in the perinatal constant light group were 2.6, 2.5 and 8.5 times higher than the LD12:12 controls, respectively. As for tumour-bearing animals, the incidences were: males (29/34 (85%) vs 16/61 (26%); P < 0.01) and females (38/54 (70%) versus 21/66 (32%); P < 0.01). Similar results were seen for the peripheral nervous tissue (males 21/34 (62%) versus 12/61 (20%); P < 0.01 and females 29/54 (54%) versus 17/66 (26%); P < 0.01) and for the kidney tumours (males 7/34 (21%) versus 1/61 (1.6%); P < 0.01 and females 5/54 (9%) versus 1/66 (1.5%); P > 0.05). All kidney tumours were classified as mesenchymal tumours with approximately equal localization to the left and right lobes. The nervous system tumours were equally distributed between benign (fascicular neurinoma, reticular neurinoma) and malignant (neuroblastoma, other malignant) [Even though this information was available, the Working Group was unable to separate malignant from benign tumours due to tumour multiplicity within animals in the same group]. The group exposed to perinatal constant darkness demonstrated a significant drop in the number of tumour-bearing animals for males (5/40 (12%); P < 0.01) and females (5/44 (11%); P < 0.01) and for both sexes combined (P < 0.01). For peripheral nervous system tumours (males 4/40 (10%); P > 0.05 and females 3/44 (7%); P < 0.05) the significant drop in tumour incidence applied only to females or both sexes combined. Kidney tumours (males 1/40 (2.5%); P > 0.05 and females 4/44 (9%); P > 0.05) showed no change from control in the perinatal constant darkness group. Finally, the mean survival of tumour-bearing rats [detected at necropsy] was significantly reduced in the constant light group (P < 0.01 for nervous system tumours, P < 0.05 for kidney tumours) for both tumour types and significantly extended (P < 0.05 for both tumour types) in the perinatal constant darkness group for both tumour types when combining both sexes (Beniashvili et al., 2001).

3.2 Effects of pinealectomy and nocturnal physiological melatonin levels on the development and/or growth of chemically induced or transplantable experimental tumours in animals

Introduction

Pinealectomy consists in the surgical removal of the pineal gland from the brain. It is the only means of eliminating the nocturnal melatonin signal emanating from the pineal gland without also affecting the central circadian pacemaker in the SCN of the hypothalamus. Pinealectomy has been employed as one means of determining whether

SHIFTWORK 649

the specific suppression of the physiological nocturnal melatonin signal leads to the enhancement of cancer development and/or growth in experimental animal models of tumorigenesis (see Table 3.1). At the same time, this procedure indirectly addresses whether the physiological nocturnal melatonin signal from the pineal gland is inhibitory to the process of tumorigenesis in experimental animal models. However, it is important to note that unidentified, non-melatonin compounds (i.e. small peptides) that possess anticancer activity both in vivo and in vitro have been isolated from the pineal gland (Bartsch et al., 1992). Therefore, the mere removal of the pineal gland in the absence of physiological melatonin replacement would not unequivocally prove that only melatonin is responsible for the antineoplastic effects of the pineal or that the promotion of tumorigenesis by pinealectomy is exclusively due to the elimination of the nocturnal physiological melatonin signal. However, in view of the fact that these putative oncostatic substances have never been structurally identified, measured in the blood or other extracellular fluids or determined to be mediators of pineal/circadian physiology, their role in the pineal regulation of tumorigenesis will not be considered. In all of the studies that follow, the melatonin levels were not measured. Other sources of melatonin have been identified in rodents, including the Harderian gland and the intestine. Their relative contribution to the physiological rhythm in circulating melatonin levels are still poorly understood. [However, the Working Group felt confident that, in these studies, pinealectomy resulted in marked reduction in melatonin levels.]

3.2.1 Undifferentiated neoplasms (Yoshida and Ehrlich tumour)

Neonatal pinealectomy 24 hours after birth was evaluated on the growth and metastatic spread of Yoshida solid tumour cells transplanted intramuscularly into Sprague-Dawley rats [sex unspecified] 10–12 weeks following pinealectomy. Survival time was decreased in pinealectomized rats (n = 10) over intact control rats (P < 0.001) (n = 4). There was no difference in tumour weight between the pinealectomized and control groups. The prevalence of tumour metastases to the pancreas was markedly increased whereas metastatic foci were much less in liver [no statistical analysis] (Lapin, 1974).

The growth and mitotic index of another undifferentiated tumour, Ehrlich tumour, intraperitoneally or subcutaneously injected into six groups of Swiss inbred mice (25 g) that were either pinealectomized or sham-operated or not operated (controls) were evaluated. [The number of animals per group was not precisely provided but was assumed to be 17–18. Because of the high mortality, the Working Group had concerns regarding the adequacy of the sample size.] Pinealectomized animals were found to have more intraperitoneal ascite tumours (P < 0.05) with a greater mitotic index (P < 0.01). Solid subcutaneous tumour weight did not change, although there was an increase in the solid tumour mitotic index (P < 0.001) (Billitteri & Bindoni, 1969).


3.2.2 Sarcoma

As early as the 1940s it was demonstrated that pinealectomy could stimulate the growth of transplantable sarcomas in rats (Nakatani et al., 1940; Katuguiri, 1943).

Thirty years later, the effects of pinealectomy were demonstrated 7 weeks following the subcutaneous injection of previously pinealectomized Holtzman rats [sex unspecified] with fibrosarcoma cells derived from rats treated with methylcholanthrene. Mean tumour volume in these animals was over 2-fold greater in pinealectomized rats than that in the intact control rats, and nearly 2-fold greater than in sham-operated rats. The prevalence of lymph node metastases was 2.5-fold (P < 0.05) [P = 0.013; pinealectomized versus sham-operated] greater in the pinealectomized group than in the combined control groups while the number of animals with lung metastases was virtually the same among all treatment groups (Barone et al., 1972). The comparison of mean tumour volume in pinealectomized rats with that in the combined controls showed a significant increase (P < 0.01). [The Working Group questioned whether to use combined intact and sham group was reasonable.]

In contrast, exposure of Wistar rats [sex unspecified] to the polyoma virus failed to induce neosarcoma in neonatally pinealectomized or intact control animals (Wrba et al., 1975). [The lack of details and of an effect in the controls make an interpretation of this study problematic.]

3.2.3 Hepatocarcinoma

Pinealectomy has been reported to inhibit the development of chemically induced hepatocarcinomas in rats (Lacassagne et al., 1969).

More recently, the effects of pinealectomy versus sham-pinealectomy were examined to evaluate the growth of transplantable tissue-isolated Morris rat hepatoma (7288CTC) in male Buffalo rats over a 2-week period. The tumour growth rate in animals that were pinealectomized (n = 8) one week before tumour implantation was 2-fold (P < 0.05) greater than the tumour growth rate in sham-pinealectomized controls (n = 8) over a 3-week period, and latency to tumour onset was reduced by 50% (P < 0.05). The tumour uptake of linoleic acid and production of its metabolite 13-HODE, the mitogenic signal upon which hepatoma 7288CTC is dependent, were markedly increased in pinealectomized rats versus their sham-pinealectomized counterparts (P < 0.05) (Blask et al., 1999).

3.2.4 Ovarian and small bowel adenocarcinoma

Previously pinealectomized (n = 12), sham-pinealectomized (n = 10) or intact (n = 10) hamsters [sex and strain not specified] were inoculated subcutaneously with ovarian tumour cells [type not specified]. The interval of time between the surgical procedures and tumour cell inoculation was not specified. Tumour growth evaluated over

SHIFTWORK 651

a 30-day period following tumour inoculation revealed that tumour volume was about 5-fold greater in pinealectomized animals versus sham-pinealectomized; tumour volumes in sham-operated and intact hamsters were virtually equivalent [no P values shown]. No significant differences were observed in the growth of small bowel adenocarcinomas in the same pinealectomized versus sham-operated animals 2 weeks following tumour cell inoculation (Das Gupta, 1968). [The Working Group cannot clearly interpret the results of this study due to lack of details.]

3.2.5 Walker 256 carcinosarcoma

The effects of pinealectomy have been determined on the growth and spread of transplantable Walker 256 carcinosarcomas (carcinomatous variant) in male Sprague-Dawley rats. Young adult rats (40–50 g; 13 rats per group) were either pinealectomized, sham-pinealectomized or left intact 2 weeks before being injected into the thigh muscle with a homogenate of Walker carcinoma. Tumour size was measured every 2 days until the occurrence of spontaneous death from the tumour. The survival time of pinealectomized rats was significantly decreased by 14.6% compared to sham-pinealectomized rats (P < 0.02). Tumour size was significantly increased in pinealectomized animals by 43% versus sham-pinealectomized animals (P < 0.01). There were a greater number of rats with lung or lymph node metastases in the pinealectomized group than in either the sham-pinealectomized or intact groups, although the statistical significance of these differences were not determined (Rodin, 1963) [A Fisher’s exact test performed by the Working Group revealed that the prevalence of nodal, but not lung, metastases in pinealectomized rats was significantly higher than in the sham-operated group (P < 0.04).]

In another confirmatory study in young inbred male Holtzman rats (40 – 60 g) (Barone & Das Gupta, 1970), it was demonstrated that the mean tumour volume in pinealectomized rats (n = 27) 24 days following subcutaneous injection of a cell suspension of Walker 256 carcinoma was 42% greater than in sham-operated animals (n = 24) (P < 0.01); pinealectomy and sham-pinealectomy were carried out 5 weeks before tumour cell injection. There was also a greater number of pinealectomized animals (n = 39) with metastatic lesions localized to the lungs as well as axillary and mediastinal lymph nodes than sham-pinealectomized rats (n = 35); however, no statistical analysis was performed [A Fisher’s exact test revealed that these differences were statistically different for axillary nodes (P < 0.02), mediastinal nodes (P < 0.001), and lung (P < 0.001).]

3.2.6 Melanoma

The effects of pinealectomy were evaluated on the growth and metastatic spread of transplantable hamster melanoma cells – Melanotic Melanoma No. 1 (MM1) – in adult male and female Syrian hamsters. Animals were inoculated with a melanoma cell


suspension 5 weeks following pinealectomy or sham-operation; an additional control group was left intact. Tumour volume was measured every week for 5 weeks following tumour cell inoculation. At the end of the first 2 weeks following tumour cell inoculation, tumour volume was 10-fold higher in the pinealectomized group (n = 10) versus the sham-pinealectomized animals (n = 10) (P < 0.001). The overall tumour growth rate over the subsequent 3 weeks in the pinealectomized group was 3-fold higher than in the sham-pinealectomized group (P < 0.001); no significant differences were observed between the sham-operated and intact controls. There was also a higher frequency of metastasic foci in the lungs, liver, kidneys, spleen and axillary lymph nodes in the pinealectomized group when compared to sham-operated animals (P < 0.001) through 21 days; no significant differences were observed between the sham-operated and intact controls (Das Gupta & Terz, 1967).

In a follow-up study, melanoma growth was examined in young adult male Syrian hamsters, 4–6 weeks of age, that were either pinealectomized or sham-pinealectomized. One week following pinealectomy or sham surgery, the animals were injected subcutaneously with a tumour suspension of MM1 hamster melanoma cells derived from solid tumour tissue, and tumour weights were evaluated 3 and 6 weeks following tumour injection. After 3 weeks, mean tumour weight in the pinealectomized animals (n = 12) was 2.3-fold higher (P < 0.05) than in sham-operated animals (n = 11), and after 6 weeks it was 1.6-fold higher (pinealectomized, n = 13; sham-operated, n = 12; P < 0.05); tumour weight was not significantly different between sham-operated and intact animals either after 3 or 6 weeks (n = 10–11) (El-Domeiri & Das Gupta, 1973).

In a later study by this group, the effects of pinealectomy versus sham-pinealectomy were evaluated on the growth of MM1 hamster melanoma growth in young adult male Syrian hamsters, 5–6 weeks of age, under the conditions of a long photoperiod (LD14:10) or short photoperiod (LD6:18). Animals were maintained on either long or short days for 2 weeks before pinealectomy or sham surgery and continued on these photoperiods thereafter. One week following surgery, animals were injected subcutaneously with a suspension of cells derived from MM1 hamster melanoma. Under long days, the tumour growth rate in the pinealectomized hamsters (n = 16) was higher than in sham-operated animals (n = 11) over 38 days as determined by serial measurement of tumour volumes (no statistical comparison); tumour latency was identical in both groups. The final mean tumour weight in pinealectomized hamsters was 37% higher than in sham-operated controls (P < 0.01). In contrast, under short days, the tumour growth rate was lower in pinealectomized hamsters (n = 8) than in sham-operated controls (n = 9) over 51 days (no statistical comparison); tumour latency was significantly longer in pinealectomized versus sham-operated animals (P < 0.05). The final mean tumour weight in pinealectomized animals was nearly 50% lower than in sham controls (P < 0.01) (Stanberry et al., 1983).

In a study by another group in a carcinogen-induced model of melanoma, the effects of pinealectomy versus sham-pinealectomy were examined on the development of melanomas induced by the intragastric administration of DMBA in male and female Syrian hamsters 2 days after pinealectomy or sham surgery. Thirteen months following

SHIFTWORK 653

DMBA administration, the number of melanomas (> 1 mm and < 5 mm) in pinealectomized male animals (n = 26) was 74% higher than in sham-operated animals (n = 45) (P < 0.001), and 44% higher in pinealectomized females (n = 11) than in sham-operated females (n = 20) (P < 0.001). No significant differences were observed in tumour number in pinealectomized versus sham-operated males or females for tumours > 5 mm although tumour number tended to be smaller in the pinealectomized groups than in the sham-operated groups. The effects of pinealectomy on mean tumour size or incidence was not determined in this study (Aubert et al., 1970). [No direct comparisons were done to compare direct measures of tumour size between groups.]

3.2.7 Prostate carcinoma

Only one study has examined the effects of pinealectomy on the growth of a fast-growing, androgen-independent transplantable rat Dunning R3327 prostate cancer in adult male Copenhagen-Fischer F1 rats. Tumours were transplanted subcutaneously into pinealectomized (n = 11) or sham-operated rats (n = 10) [timing of surgery and tumour implantation in relationship to surgery were not specified]; no differences in growth rates were observed over a 75-day period following tumour transplantation (Toma et al., 1987).

3.2.8 Uterine carcinoma (Guerin malignant epithelioma)

DMBA Guerin malignant spontaneous epitheliomas of the Wistar rat uterus were transplanted

into adult male Wistar rats that were either pinealectomized (n = 7) or left intact (n = 17). Three days after the pinealectomy, all rats were transplanted subcutaneously with Guerin epitheliomas. Mean life span in tumour-bearing pinealectomized animals was significantly reduced by 14 days when compared to intact controls (P < 0.001); the mean mitotic activity of tumours in pinealectomized rats was moderately (15%) but significantly higher than in intact controls (P < 0.05) 35–45 days after tumour transplantation (Lewiński et al., 1993).

3.2.9 Mammary carcinoma

The effects of neonatal pinealectomy (24 hours after birth) on the development of mammary tumours induced by the intragastric administration of DMBA in adult female Wistar rats were evaluated against intact animals. Both groups of animals received a total of three intragastric DMBA treatments separated by 10-day intervals 3–3.5 months after pinealectomy, and were treated with saline following the first DMBA treatment. The incidence and final prevalence of DMBA-induced mammary tumours was the same in both pinealectomized (n = 12) and intact animals (n = 12) over the 400-day period following the first DMBA administration (Lapin, 1978).


In a subsequent study, the effects of pinealectomy versus sham-pinealectomy in animals on an LD12:12 light dark cycle were examined on DMBA-induced mammary tumorigenesis in young adult female Sprague-Dawley rats (58 days of age). Pinealectomy and sham surgery were performed 2 days before the administration of DMBA. Latency to onset and incidence of mammary tumours in pinealectomized rats (n = 17) versus sham-operated animals (n = 20) were not statistically different during the 140 days of tumorigenesis (Aubert et al., 1980).

Pinealectomized young adult (50 days of age) female Sprague-Dawley rats administered a single low dose of DMBA (7 mg) 30 days following surgery showed a 4-fold higher incidence of mammary tumours in the pinealectomized group relative to the sham-operated controls 240 days following DMBA treatment (P < 0.002). When pinealectomized and sham-pinealectomized rats were administered a higher dose of DMBA (10 mg), tumour development in pinealectomized rats (n = 30) was 2-fold higher than in the sham-pinealectomized group (n = 30, P < 0.03) (Tamarkin et al., 1981).

In a series of publications from one study that addressed the effects of neonatal (2 days of age) pinealectomy or sham surgery on DMBA-induced (Day 55) mammary tumorigenesis in adult female Holtzman rats maintained on a short photoperiod (LD10:14 light–dark cycle) from birth, no significant differences were found in the final prevalence of mammary tumours, mammary tumour number, mean latency to tumour onset, [3H]thymidine incorporation into DNA in mammary tissue or the number of terminal end or alveolar buds in pinealectomized rats (n = 23) versus sham-operated rats (n = 15). The total duration of tumour-monitoring was 180 days after DMBA administration. No tumour incidence curves were presented in either of these studies so that the rates of tumour development could be statistically compared (Kothari et al., 1984; Shah et al., 1984).

In a subsequent study, it was similarly demonstrated that the incidence, final (55 days of age) prevalence, and number of DMBA-induced mammary tumours in adult female Holtzman rats that undergone pinealectomy neonatally (2 days of age) were not significantly different at the end of the tumorigenic period (30 weeks post-DMBA) between neonatally pinealectomized (n = 20) and intact animals (n = 20) maintained on a short photoperiod (LD10:14). However, 80% tumour prevalence in pinealectomized animals was achieved 12 weeks following DMBA treatment when compared to intact control animals that had a 10% prevalence at 12 weeks and a maximal tumour prevalence that was not apparent until 24 weeks following DMBA administration (70% tumour prevalence). [It was clear that under short photoperiod conditions, mammary tumours in pinealectomized animals developed at a rate that was substantially faster than in intact animals; however, these investigators did not statistically analyse the tumour incidence curves presented in their report.] (Subramanian and Kothari, 1991).

In another carcinogen-induced mammary tumour model, pinealectomized (3 days before the first NMU injection) adult female Sprague-Dawley rats (n = 11) maintained on an LD12:12 light–dark cycle and treated with the carcinogen NMU, on Day 50 and Day 57, exhibited a trend for an overall increase in the incidence and number of mammary

SHIFTWORK 655

tumours over intact animals (n = 14) for the period encompassing 19 weeks following NMU injection; however, this increase was not statistically significant. Similarly, no significant difference was observed in tumour latency in pinealectomized rats versus intact controls (Blask et al., 1991).

In a subsequent study using adult female Fischer 344/N rats maintained on an LD12:12 photoperiod, animals were pinealectomized (n = 40) at 4 weeks of age or left intact (n = 40) and given NMU intraperitoneally (50 days of age). Tumour development was documented over a 26-week period following NMU administration. There were no significant differences in tumour incidence, final prevalence, number, size or latency between the pinealectomized and intact groups even though the circadian melatonin rhythm was fully expressed in the intact animals and completely extinguished in the pinealectomized rats during the first half of the study. However, by the end of the study, a nocturnal melatonin signal was present in pinealectomized rats (Travlos et al., 2001). [This latter result was difficult to explain; one possibility was that an extrapineal source of circulating melatonin (i.e. the gut) may have compensated for the loss of the pineal gland.]

3.3 Effects of physiological melatonin administration on experimental tumour growth activity in animals

Introduction

Most of the studies have demonstrated an oncostatic action of melatonin on tumour development and growth in experimental animal models of cancer. However, these studies have been performed using pharmacological doses of melatonin. The nocturnal, physiological blood concentrations of melatonin in vivo are inhibitory to tumorigenesis and have been inferred from studies employing pinealectomy as a technique for specifically eliminating the nocturnal melatonin signal and observing a stimulation of tumour development and growth (see section 3.2 above). Only a handful of recent studies (see Table 3.1) have directly investigated the role of physiological, nocturnal concentrations of melatonin on experimental cancer growth in vivo.

3.3.1 Rat hepatoma

In one study, 3-mm3 of tissue-isolated Morris rat hepatomas (7288CTC) were sutured to the tip of a vascular stalk formed from the superficial epigastric artery and vein of groups of 5–9 male Buffalo rats. When these tumours reached approximately 5 g, the carotid artery and tumour vein were either cannulated or perfused in situ. Perfusion experiments used rat whole blood harvested during the early light phase just a few hours following light onset when endogenous melatonin levels were low. Perfusion studies using a high physiological concentration of melatonin (1 nM) for 2.5 hours reversibly


(P < 0.05) blocked the uptake of linoleic acid, production of 13-HODE, and significantly decreased (P < 0.05) the incorporation of [3H]thymidine into DNA. [These findings were strengthened by studies using several melatonin receptor antagonists that completely reversed the effects of the melatonin.] In the cannulated animals measured every 4 hours for 24 hours, tumour linoleic acid uptake and metabolism to 13-HODE were temporally correlated (P < 0.05) with the circadian rhythm and a significant difference was demonstrated between peak dark and peak light phase values (P < 0.05). Finally, pinealectomized rats hosting tumour tissue and given either daily subcutaneous injections of melatonin or provided oral melatonin (200 µg) demonstrated a significant delay in the latency to tumour palpability when compared to appropriate sham controls (P < 0.05) (Blask et al., 1999).

In a subsequent dose–response study using the same experimental protocol, increasing concentrations of exogenous melatonin were administered to perfused in-situ tissue-isolated Morris rat hepatomas hosted in Buffalo rats, 4–5 weeks of age, using whole blood harvested from pinealectomized Sprague-Dawley donor rats. Final whole blood concentrations of melatonin reproduced those levels characteristic of the ascending limb of the nocturnal, endogenous melatonin surge. A significant (P < 0.05) dose-dependent suppression of tumour linoleic acid uptake and 13-HODE production occurred following melatonin perfusion. Similarly, a dose-dependent suppression of tumour-DNA content (P < 0.05) and [3H]thymidine incorporation into DNA (P < 0.05) was seen in response to melatonin along this concentration range as well. The inhibitory effects of melatonin on tumour linoleic acid metabolism and cellular replication activity saturated at the highest physiological concentration of 1 nM. Additionally, tumour uptake and retention of melatonin itself, as a function of supply, ranged from 20 to 45% across all concentrations tested. In the same study, melatonin was added to a semi-purified 5% corn oil diet so that pineal gland intact animals ingested, primarily during the dark phase, either 50 ng, 500 ng or 5 μg/day of additional dietary melatonin to produce physiological, nocturnal concentrations of melatonin that added to the endogenous nocturnal surge. When animals began receiving melatonin in their diet 2 weeks before tumour implantation and continuously thereafter, tumour growth as well as linoleic acid uptake and metabolism to 13-HODE were significantly inhibited (P < 0.05) in a dose-dependent manner (Blask et al., 2004).

3.3.2 Human cancer xenograft

Adult male Buffalo rats were implanted with 7288CTC hepatoma cells as described above (positive control), and adult female nude rats were implanted with tissue-isolated steroid receptor negative (SR−, no estrogen or progesterone receptor expression) or steroid-receptor positive (SR+) human breast cancer xenografts. The SR+ xenografts when perfused with Sprague-Dawley rat donor whole blood to which was added 1 nmol/L of synthetic melatonin showed significant reduction in [3H]thymidine incorporation (P < 0.05) and in camp levels (P < 0.05). This reduction was completely

SHIFTWORK 657

eliminated by coperfusion with 13-HODE. Although not shown, similar findings were reported for the SR– xenographs. The authors further investigated these models using differing levels of light at night to control melatonin levels. Using six animals per group and six different light intensities resulted in significant changes (P < 0.05) in Phos, ERK1/2, linoleic acid uptake, 13-HODE, c-AMP, [3H]thymidine incorporation and tumour onset in both the 7288CTC model and the SR– model. Similarly, tissue-isolated SR–, MT1 melatonin-receptor positive MCF-7 human breast cancer xenografts were perfused in situ for 1 hour with human whole blood from premenopausal females collected in daytime, night time and after exposure to bright light at night. There was a significant reduction of [3H]thymidine incorporation (63% to 73%) between samples perfused at night time and daytime (P < 0.05) which was eliminated in experiments using blood from volunteers exposed to bright light at night. Other markers as noted above (linoleic acid uptake, 13-HODE, cAMP) behaved as expected. Finally, to determine if this was entirely driven by melatonin, 500 pmol/L of melatonin was added to blood from donors exposed to bright light at night for 90 minutes. The results were identical to what was seen for the night time blood sample experiments and these results could easily be blocked by using MT1/MT2 antagonists. Using a new perfusion system that minimized delivery time, it was subsequently demonstrated that perfusion of tissue-isolated estrogen-receptor negative MCF-7 human breast cancer xenografts in situ with melatonin-containing (1 nM final concentration) daytime-collected rat donor blood completely suppressed (P < 0.05) linoleic acid uptake and 13-HODE formation within 5 minutes of melatonin reaching the tumour, indicating that the melatonin suppression of tumour linoleic acid metabolism is extremely rapid (Dauchy et al., 2006).

Using the same basic protocol used above, tissue-isolated FaDu human squamous-cell cancer xenografts (grade II human hypopharyngeal squamous cell carcinoma) were implanted into male athymic nude rats. Cancer xenografts were perfused in situ for 2 hours with daytime-collected male adult Buffalo rat donor whole blood to which melatonin was added at a final concentration of 1 nM. This perfusion resulted in a total blockade of linoleic acid uptake (P < 0.05) and 13-HODE formation (P < 0.05) as well as a significant (P < 0.05) 76% suppression of cAMP levels and a 50% inhibition (P < 0.05) of [3H]thymidine incorporation into DNA and DNA content when compared to vehicle-containing daytime-collected control whole blood (Dauchy et al., 2007).

3.4 References

Anderson LE, Morris JE, Sasser LB, Stevens RG (2000). Effect of constant light on DMBA mammary tumorigenesis in rats. Cancer Lett, 148:121–126 doi:10.1016/S0304-3835(99)00320-1. PMID:10695987

Anisimov VN, Baturin DA, Popovich IG et al. (2004). Effect of exposure to light-at-night on life span and spontaneous carcinogenesis in female CBA mice. Int J Cancer, 111:475–479 doi:10.1002/ijc.20298. PMID:15239122


Aubert C, Janiaud P, Lecalvez J (1980). Effect of pinealectomy and melatonin on mammary tumor growth in Sprague-Dawley rats under different conditions of lighting. J Neural Transm, 47:121–130 doi:10.1007/BF01670163. PMID:6768843

Aubert C, Prade M, Bohuon C (1970). [Effect of pinealectomy on the melanic tumours of the golden hamster induced by administration (per os) of a single dose of 9,10-dimethyl-1,2-benzanthracene]. C R Acad Sci Hebd Seances Acad Sci D, 271:2465–2468. PMID:4995225

Balsalobre A, Brown SA, Marcacci L et al. (2000). Resetting of circadian time in peripheral tissues by glucocorticoid signaling. Science, 289:2344–2347 doi:10.1126/science.289.5488.2344. PMID:11009419

Barone RM, Abe R, Das Gupta TK (1972). Pineal ablation in methylcholanthrene-induced fibrosarcoma. Surg Forum, 23:115–116. PMID:4671044

Barone RM, Das Gupta TK (1970). Role of pinealectomy on Walker 256 carcinoma in rats. J Surg Oncol, 2:313–322 doi:10.1002/jso.2930020402. PMID:5520847

Bartsch H, Bartsch C, Simon WE et al. (1992). Antitumor activity of the pineal gland: effect of unidentified substances versus the effect of melatonin. Oncology, 49:27–30 doi:10.1159/000227005. PMID:1542489

Beniashvili DS, Benjamin S, Baturin DA, Anisimov VN (2001). Effect of light/dark regimen on N-nitrosoethylurea-induced transplacental carcinogenesis in rats. Cancer Lett, 163:51–57 doi:10.1016/S0304-3835(00)00673-X. PMID:11163108

Billitteri A, Bindoni M (1969). [Growth and cell multiplication of Ehrlich’s tumor in the mouse, after removal of the pineal body]. Boll Soc Ital Biol Sper, 45:1647–1650. PMID:5400878

Blask DE, Brainard GC, Dauchy RT et al. (2005). Melatonin-depleted blood from premenopausal women exposed to light at night stimulates growth of human breast cancer xenografts in nude rats. Cancer Res, 65:11174–11184 doi:10.1158/0008-5472.CAN-05-1945. PMID:16322268

Blask DE, Dauchy RT, Sauer LA et al. (2003). Growth and fatty acid metabolism of human breast cancer (MCF-7) xenografts in nude rats: impact of constant light-induced nocturnal melatonin suppression. Breast Cancer Res Treat, 79:313–320 doi:10.1023/A:1024030518065. PMID:12846415

Blask DE, Dauchy RT, Sauer LA, Krause JA (2004). Melatonin uptake and growth prevention in rat hepatoma 7288CTC in response to dietary melatonin: melatonin receptor-mediated inhibition of tumor linoleic acid metabolism to the growth signaling molecule 13-hydroxyoctadecadienoic acid and the potential role of phytomelatonin. Carcinogenesis, 25:951–960 doi:10.1093/carcin/bgh090. PMID:14754876

Blask DE, Pelletier DB, Hill SM et al. (1991). Pineal melatonin inhibition of tumor promotion in the N-nitroso-N-methylurea model of mammary carcinogenesis: potential involvement of antiestrogenic mechanisms in vivo. J Cancer Res Clin Oncol, 117:526–532 doi:10.1007/BF01613283. PMID:1744157

Blask DE, Sauer LA, Dauchy RT et al. (1999). Melatonin inhibition of cancer growth in vivo involves suppression of tumor fatty acid metabolism via melatonin receptor-mediated signal transduction events. Cancer Res, 59:4693–4701. PMID:10493527

Clausen OP, Iversen OH, Thorud E (1984). Circadian variation in the susceptibility of mouse epidermis to tumour induction by methylnitrosourea. Virchows Arch B Cell Pathol Incl Mol Pathol, 45:325–329 doi:10.1007/BF02889874. PMID:6146223

SHIFTWORK 659

Cos S, Mediavilla D, Martínez-Campa C et al. (2006). Exposure to light-at-night increases the growth of DMBA-induced mammary adenocarcinomas in rats. Cancer Lett, 235:266–271 doi:10.1016/j.canlet.2005.04.025. PMID:15950374

Das Gupta TK (1968). Influence of the pineal gland on the growth and spread of malignant tumors. Surg Forum, 19:83–84. PMID:5752735

Das Gupta TK, Terz J (1967). Influence of pineal gland on the growth and spread of melanoma in the hamster. Cancer Res, 27:1306–1311. PMID:4952523

Dauchy EM, Dauchy RT, Davidson LK et al. (2006). Human cancer xenograft perfusion in situ in rats: a new perfusion system that minimizes delivery time and maintains normal tissue physiology and responsiveness to growth-inhibitory agents. J Am Assoc Lab Anim Sci, 45:38–44. PMID:16642969

Dauchy RT, Dauchy EM, Davidson LK et al. (2007). Inhibition of fatty acid transport and proliferative activity in tissue-isolated human squamous cell cancer xenografts perfused in situ with melatonin or eicosapentaenoic or conjugated linoleic acids. Comp Med, 57:377–382. PMID:17803052

Dauchy RT, Sauer LA, Blask DE, Vaughan GM (1997). Light contamination during the dark phase in “photoperiodically controlled” animal rooms: effect on tumor growth and metabolism in rats. Lab Anim Sci, 47:511–518. PMID:9355094

Deprés-Brummer P, Bourin P, Pages N et al. (1997). Persistent T lymphocyte rhythms despite suppressed circadian clock outputs in rats. Am J Physiol, 273:R1891–R1899. PMID:9435642

Deprés-Brummer P, Lévi F, Metzger G, Touitou Y (1995). Light-induced suppression of the rat circadian system. Am J Physiol, 268:R1111–R1116. PMID:7771569

Dubina MV, Petrishev NN, Panchebko AV et al. (2002). [Circadian features of carcinogenesis of the large intestine induced by 1,2-dimethylhydrazine in rats]. Vopr Onkol, 48:331–334. PMID:12455357

el-Domeiri AA, Das Gupta TK (1973). Reversal by melatonin of the effect of pinealectomy on tumor growth. Cancer Res, 33:2830–2833. PMID:4748439

Fan Y, Hida A, Anderson DA et al. (2007). Cycling of CRYPTOCHROME proteins is not necessary for circadian-clock function in mammalian fibroblasts. Curr Biol, 17:1091–1100 doi:10.1016/j.cub.2007.05.048. PMID:17583506

Filipski E, Delaunay F, King VM et al. (2004). Effects of chronic jet lag on tumor progression in mice. Cancer Res, 64:7879–7885 doi:10.1158/0008-5472.CAN-04-0674. PMID:15520194

Filipski E, Innominato PF, Wu M et al. (2005). Effects of light and food schedules on liver and tumor molecular clocks in mice. J Natl Cancer Inst, 97:507–517 doi:10.1093/jnci/dji083. PMID:15812076

Filipski E, King VM, Li X et al. (2002). Host circadian clock as a control point in tumor progression. J Natl Cancer Inst, 94:690–697. PMID:11983758

Fu L, Lee CC (2003). The circadian clock: pacemaker and tumour suppressor. Nat Rev Cancer, 3:350–361 doi:10.1038/nrc1072. PMID:12724733

Fu L, Pelicano H, Liu J et al. (2002). The circadian gene Period2 plays an important role in tumor suppression and DNA damage response in vivo. Cell, 111:41–50 doi:10.1016/S0092-8674(02)00961-3. PMID:12372299

Gauger MA, Sancar A (2005). Cryptochrome, circadian cycle, cell cycle checkpoints, and cancer. Cancer Res, 65:6828–6834 doi:10.1158/0008-5472.CAN-05-1119. PMID:16061665


Hamilton T (1969). Influence of environmental light and melatonin upon mammary tumour induction. Br J Surg, 56:764–766 doi:10.1002/bjs.1800561018. PMID:5347580

Harmar AJ, Marston HM, Shen S et al. (2002). The VPAC(2) receptor is essential for circadian function in the mouse suprachiasmatic nuclei. Cell, 109:497–508 doi:10.1016/S0092-8674(02)00736-5. PMID:12086606

Hastings MH, Reddy AB, Maywood ES (2003). A clockwork web: circadian timing in brain and periphery, in health and disease. Nat Rev Neurosci, 4:649–661 doi:10.1038/nrn1177. PMID:12894240

Ishida A, Mutoh T, Ueyama T et al. (2005). Light activates the adrenal gland: timing of gene expression and glucocorticoid release. Cell Metab, 2:297–307 doi:10.1016/j.cmet.2005.09.009. PMID:16271530

Iversen OH, Iversen UM (1995). A diurnal variation in the tumorigenic response of mouse epidermis to a single application of the strong short-acting chemical carcinogen methylnitrosourea. A dose-response study of 1, 2 and 10 mg. In Vivo, 9:117–132. PMID:7548787

Jöchle W (1963) Wirkungen von Dauerbelichtung und Sulfonamidverabreichung auf Cyclus und spontanes Mammatumorwachstum bei 2 C3H Mäuse-Umzuchstämmen. Symposium, Deutsche Gesellschaft fur Endokrinologie Symposion 10: 305–308.

Joechle W (1964). Trends in photophysiologic concepts. Ann N Y Acad Sci, 117:88–104 doi:10.1111/j.1749-6632.1964.tb48164.x. PMID:14196671

Katuguiri E (1943) Studies on the pineal gland. Tumor proliferation and the pineal gland. Osaka Igakkai Zasshi, 42:935–938.

Khaetski IK (1965). [Effect of hypothalamo-pituitary lesions induced by constant illumination on development of induced mammary tumors in rats]. Vopr Exp Oncol (Kiev), 1:87–93.

Kothari LS, Shah PN, Mhatre MC (1982). Effect of continuous light on the incidence of 9,10-dimethyl-1,2-benzanthracene induced mammary tumors in female Holtzman rats. Cancer Lett, 16:313–317 doi:10.1016/0304-3835(82)90012-X. PMID:6817910

Kothari LS, Shah PN, Mhatre MC (1984). Pineal ablation in varying photoperiods and the incidence of 9,10-dimethyl-1,2-benzanthracene induced mammary cancer in rats. Cancer Lett, 22:99–102 doi:10.1016/0304-3835(84)90050-8. PMID:6421481

Lacassagne A, Chamorro A, Hurst L, Nguyen-Ba-Giao (1969). [Effect of epiphysectomy on chemical hepatocancerogenesis in rats]. C R Acad Sci Hebd Seances Acad Sci D, 269:1043–1046. PMID:4981428

Lapin V (1974). Influence of simultaneous pinealectomy and thymectomy on the growth and formation of metastases of the Yoshida sarcoma in rats. Exp Pathol (Jena), 9:108–112. PMID:4452369

Lapin V (1978). Effects of reserpine on the incidence of 9,10-dimethyl-1,2-benzanthracene-induced tumors in pinealectomised and thymectomised rats. Oncology, 35:132–135 doi:10.1159/000225271. PMID:97606

Levi F, Schibler U (2007). Circadian rhythms: mechanisms and therapeutic implications. Annu Rev Pharmacol Toxicol, 47:593–628 doi:10.1146/annurev.pharmtox.47.120505.105208. PMID:17209800

Lewiński A, Sewerynek E, Wajs E et al. (1993). Effects of the pineal gland on the growth processes of Guerin epithelioma in male Wistar rats. Cytobios, 73:89–94. PMID:8319500

SHIFTWORK 661

Mhatre MC, Shah PN, Juneja HS (1984). Effect of varying photoperiods on mammary morphology, DNA synthesis, and hormone profile in female rats. J Natl Cancer Inst, 72:1411–1416. PMID:6427503

Nagano M, Adachi A, Nakahama K et al. (2003). An abrupt shift in the day/night cycle causes desynchrony in the mammalian circadian center. J Neurosci, 23:6141–6151. PMID:12853433

Nagashima K, Matsue K, Konishi M et al. (2005). The involvement of Cry1 and Cry2 genes in the regulation of the circadian body temperature rhythm in mice. Am J Physiol Regul Integr Comp Physiol, 288:R329–R335. PMID:15331384

Nakatani M, Ohara Y, Katagiri E et al. (1940). Studies on pinealectomized white rat. Nippon Byori Gakkai Kaishi, 30:232–236.

Pereira MA, Barnes LH, Rassman VL et al. (1994). Use of azoxymethane-induced foci of aberrant crypts in rat colon to identify potential cancer chemopreventive agents. Carcinogenesis, 15:1049–1054 doi:10.1093/carcin/15.5.1049. PMID:8200067

Reddy AB, Field MD, Maywood ES, Hastings MH (2002). Differential resynchronisation of circadian clock gene expression within the suprachiasmatic nuclei of mice subjected to experimental jet lag. J Neurosci, 22:7326–7330. PMID:12196553

Rodin AE (1963). The Growth and spread of walker 256 carcinoma in pinealectomized rats. Cancer Res, 23:1545–1548. PMID:14072694

Schibler U, Brown SA (2005). Enlightening the adrenal gland. Cell Metab, 2:278–281 doi:10.1016/j.cmet.2005.10.001. PMID:16271527

Shah PN, Mhatre MC, Kothari LS (1984). Effect of melatonin on mammary carcinogenesis in intact and pinealectomized rats in varying photoperiods. Cancer Res, 44:3403–3407. PMID:6430548

Skene DJ, Lockley SW, Thapan K, Arendt J (1999). Effects of light on human circadian rhythms. Reprod Nutr Dev, 39:295–304 doi:10.1051/rnd:19990302. PMID:10420432

Stanberry LR, Das Gupta TK, Beattie CW (1983). Photoperiodic control of melanoma growth in hamsters: influence of pinealectomy and melatonin. Endocrinology, 113:469–475 doi:10.1210/endo-113-2-469. PMID:6872938

Subramanian A, Kothari L (1991). Suppressive effect by melatonin on different phases of 9,10-dimethyl-1,2-benzanthracene (DMBA)-induced rat mammary gland carcinogenesis. Anticancer Drugs, 2:297–303 doi:10.1097/00001813-199106000-00013. PMID:1802026

Tamarkin L, Cohen M, Roselle D et al. (1981). Melatonin inhibition and pinealectomy enhancement of 7,12-dimethylbenz(a)anthracene-induced mammary tumors in the rat. Cancer Res, 41:4432–4436. PMID:6796259

Toma JG, Amerongen HM, Hennes SC et al. (1987). Effects of olfactory bulbectomy, melatonin, and/or pinealectomy on three sublines of the Dunning R3327 rat prostatic adenocarcinoma. J Pineal Res, 4:321–338 doi:10.1111/j.1600-079X.1987.tb00870.x. PMID:3625464

Travlos GS, Wilson RE, Murrell JA et al. (2001). The effect of short intermittent light exposures on the melatonin circadian rhythm and NMU-induced breast cancer in female F344/N rats. Toxicol Pathol, 29:126–136 doi:10.1080/019262301301418937. PMID:11215676

van den Heiligenberg S, Deprés-Brummer P, Barbason H et al. (1999). The tumor promoting effect of constant light exposure on diethylnitrosamine-induced hepatocarcinogenesis in rats. Life Sci, 64:2523–2534 doi:10.1016/S0024-3205(99)00210-6. PMID:10403512

Wille JJ Jr (2003). Circadian rhythm of tumor promotion in the two-stage model of mouse tumorigenesis. Cancer Lett, 190:143–149 doi:10.1016/S0304-3835(02)00594-3. PMID:12565168


Wrba H, Lapin V, Dostal V (1975). The influence of pinealectomy and of pinealectomy combined with thymectomy oncogenesis caused by polyoma virus in rats. Osterr Z Onkol, 2:37–39. PMID:174046

SHIFTWORK 663

4. Mechanistic and Other Relevant Data

4.1 The pineal gland and melatonin

Introduction

There is considerable interest in the role of the pineal gland in the development and growth of malignant tumours and in the ability of melatonin, its main secretory product, to act as an oncostatic agent.

Melatonin is a messenger of time in the mammalian organism which transmits the information of environmental light and darkness obtained from the eye through the hypothalamus to all tissues of the body. It interacts with the mechanisms that form the mammalian time structure, and has to be understood in its relation to the organism’s biological clock.

Melatonin has anti-proliferative effects on human cancer cells cultured in vitro. These oncostatic effects have been observed at physiological concentrations, and include reduction of cell-cycle progression by increasing the expression of the tumour-suppressor gene TP53, and inhibition of DNA synthesis. In addition, melatonin reduces the invasive and metastatic properties of human cancer cells in vitro, and increases intercellular communication between these cells. There is evidence from animal models that melatonin inhibits or reduces the induction of DNA damage by free radicals. Pinealectomized rats showed a higher level of DNA damage in response to treatment with a carcinogen than did rats with intact pineal glands. Melatonin also upregulates anti-oxidant enzyme systems.

4.1.1 The pineal gland and its innervations

The mammalian pineal is a secretory organ with specialized glandular cells, the pinealocytes, interstitial glial cells, and perivascular macrophages. The principal innervation is sympathetic and arises from the superior cervical ganglion. In addition, parasympathetic, commissural and peptidergic innervation are present. The sympathetic fibres contain norepinephrine and neuropeptide Y as neurotransmitters. The parasympathetic fibres contain vasoactive intestinal peptide and peptide histidine isoleucine. Neurons from the trigeminal ganglion reach the pineal gland containing substance P, calcitonin-gene-related peptide, and pituitary adenyl-cyclase-related activating peptide. Through the pineal stalk, nerve fibres originating in the brain and containing a variety of neurotransmitters innervate the pineal. In addition to its principal noradrenergic innervation, numerous receptors have been found in the pinealocyte cell


membrane, which are able to bind numerous neurotransmitters and influence the pinealocyte (for a review, see Møller & Baeres, 2002).

The secretory products of the pineal consist of melatonin which plays a major role in plant, animal, and human physiology, and several peptides the action of which is less well characterized.

4.1.2 Melatonin and its production

Melatonin (N-acetyl-5-methoxytryptamine) was first isolated by Lerner et al. (1958) from bovine pineal glands. Tryptophan is taken up from the blood stream and transformed to melatonin in four successive intracellular steps which are catalysed by tryptophan hydroxylase (EC1.14.16.4), aromatic amino acid decarboxylase (EC 4.1.1.28), arylalkylamine-N-acetyltransferase (EC 2.3.1.87), and hydroxyindole-O-methyltransferase (EC 2.1.1.4) (Axelrod & Weissbach, 1960; Lovenberg et al., 1967). Arylalkylamine-N-acetyltransferase is thought to be the rate-limiting enzyme in the process of melatonin synthesis (Klein, 2007).

Melatonin is present in bacteria, in eukaryotic unicells, in numerous plants, vegetables, fruits, seeds, rice, wheat, and medicinal herbs, and diverse species of invertebrates (Hardeland & Poeggeler, 2003).

4.1.3 Extrapineal production of melatonin

The main source of circulating melatonin in mammals is the pineal gland. However, many extrapineal mammalian tissues and organs have the enzymatic mechanism to produce melatonin (Carrillo-Vico et al., 2004, 2005).

The apparently large quantity of melatonin produced in tissues other than the pineal gland does not appear to contribute substantially to the circadian-rhythm-related plasma melatonin concentration as suppression of pineal function induced by surgical pinealectomy or constant light exposure markedly diminishes the circulating melatonin concentration (e.g. in Syrian hamsters) (Vaughan & Reiter 1986) and eliminates the nocturnal plasma melatonin concentration surge.

4.1.4 Pineal production of melatonin

In contrast, melatonin production in the pineal region is, in all mammalian species, periodic with high values during the dark phase irrespective of the activity or rest span of the species studied. In darkness, a marked 7–150 fold increase of arylalkylamine-N-acetyltransferase activity has been measured in the pineal region. The rhythm of production is endogenously generated by the activity of the suprachiasmatic nucleus (SCN) in the hypothalamus by a closed loop negative feedback of clock-gene expression. The rhythm is synchronized primarily by the environmental light–dark cycle. With irregular schedules of light exposure, it may be altered in its timing and in the duration of melatonin production. The pineal melatonin rhythm is driven by the circadian clock in the

SHIFTWORK 665

hypothalamus through a multisynaptic pathway which consists of ganglion cells in the retina containing non-vision-dependent photoreceptors (Lucas et al., 1999; Freedman et al., 1999). Retinal ganglion cells containing the pigment melanopsin are thought to be the photoreceptors for the photic entrainment of circadian rhythms (Berson et al., 2002). The photic information is transmitted to the central pacemaker in the SCN via the retino-hypothalamic tract.

The neural pathway from the SCN to the pineal region passes through pre-autonomic neurons of the paraventricular nucleus of the hypothalamus through the upper part of the spinal cord where synaptic connections are made with sympathetic preganglionic neurons relating to the superior sympathetic ganglia of the sympathetic chain (Møller & Baeres 2002). From there, postganglionic noradrenergic sympathetic neurons extend to the pineal gland. Norepinephrine is the neurotransmitter stimulating melatonin release in the pineal gland. It is released during the daily dark span in response to stimulating signals from the SCN.

Once formed, melatonin is not stored within the pineal gland but diffuses out into the capillary blood and cerebrospinal fluid (Tricoire et al., 2003). The half-life of melatonin is bi-exponential with a first distribution half-life of 2 minutes and a second of 20 minutes or longer (Claustrat et al., 2005). Melatonin released into the cerebrospinal fluid via the pineal recess reaches in the third ventricle concentrations up to 20–30 times higher than in the blood but decreases in concentration with increasing distance from the pineal region, suggesting cerebral tissue uptake (Tricoire et al., 2003).

4.1.5 Metabolism of melatonin

Circulating melatonin is metabolized mainly in the liver which clears over 90% of the circulating melatonin. Melatonin is first hydroxylated and then conjugated as sulfate and excreted as 6-sulfatoxymelatonin (aMT6s) and in a small amount as glucuronide (Arendt, 1995; Claustrat et al., 2005). Urinary and salivary aMT6s excretion closely parallels the plasma aMT6s profile. Melatonin passes into saliva in a low concentration which in its relative amount and timing corresponds also to the plasma profile. Urinary aMT6s and salivary melatonin lend themselves to the non-invasive study of melatonin secretion and its timing (Nowak et al., 1987; Arendt, 1995).

Melatonin can also be metabolized non-enzymatically in cells and extracellularly by free radicals and other oxidants. Owing to its lipophilic character, it diffuses through cell membranes easily and can exert not only receptor-dependent but also receptor-independent actions (Claustrat et al., 2005).

Once released into the blood, melatonin is primarily bound to albumin (70%) (Cardinali et al., 1972), α-1-acid glycoprotein (Morin et al., 1997), and haemoglobin (Gilad & Zisapel, 1995). Changes in circulating melatonin levels may be due in part to changes in the concentrations of one or more of these binding proteins. This could influence the availability of melatonin to various target tissues and its bioactivity in these tissues (Di et al., 1998).


4.1.6 The circadian rhythm of melatonin in plasma in humans

The plasma melatonin rhythm in humans develops between the 2nd and 3rd month of life with peak melatonin concentrations found in prepubertal children (Waldhauser et al., 1988). In the elderly, the nocturnal melatonin concentrations decrease at the end of the 5th decade and beyond to levels between 20–80% of levels found in young adults (Touitou et al., 1984; Ferrari et al., 1995; Magri et al., 1997). This drop was not found in all studies probably due to different lifestyle, state of health, sampling techniques, etc (Kennaway et al., 1999; Zeitzer et al., 1999). The circadian melatonin profile varies considerably between clinically healthy human subjects, with some having no detectable melatonin concentrations during daytime or night time (Arendt, 1985). On the other hand, in healthy individuals, the timing, amplitude, and even the shape of the melatonin profile can be highly reproducible and characteristic for a given person (Arendt, 1988; Klerman et al., 2002). Individual living habits like ‘morningness’ and ‘eveningness’ are expressed in differences in the phase of the circadian melatonin profile (Duffy et al., 1999; Gibertini et al., 1999). No consistent gender difference has been found in regard to melatonin concentrations, the melatonin profile, or its suppression by light (Arendt, 1985).

4.1.7 Light and regulation of pineal melatonin production in humans and animals

The light–dark cycle is the main entraining agent (“zeitgeber,” synchronizer) of the regulating system of pineal melatonin secretion. Light suppresses melatonin secretion in humans (Lewy et al., 1980) in an intensity-related manner (Bojkowski et al., 1987a; McIntyre et al., 1989). The endogenous rhythm governing melatonin synthesis and release is entrained to the daily dark span in different mammalian systems irrespective of the diurnal or nocturnal activity of the species. In diurnally active human subjects, high values of melatonin concentrations are released during the night. Light, in addition to acting as entraining agent of the circadian clock, will act as a masking agent when the subject is exposed to light during the habitual dark span. The photoreceptor system involved in clock regulation is distinct from the pathways associated with image formation. The sensitivity of the circadian system to light entrainment does not depend upon rod and cone photoreceptor integrity, and/or the loss of visual function (Foster et al., 1991). Eye loss in mammals, including humans, confirms that photoentrainment originates within the eye (Haus et al., 1967; Lockley et al., 1997; Foster, 1998). The daily alteration between light and dark entrains the endogenous circadian clock in the hypothalamus to the astronomical day length (24 hours). The innate period of the hypothalamic clock tends to be slightly longer than 24 hours. Absence of this input due to loss of the eye leads to a tendency of the organism to “free-run” from the 24-hour environment following the non-24-hour endogenous period of the hypothalamic clock (Haus et al., 1967; Lockley et al., 1997; Skene et al., 1999).

The photoreception acting as synchronizer upon the SCN is based upon a population of about 1% of retinal ganglion cells, which are photosensitive and respond to light directly (Sekaran et al., 2003). These photoreceptors contain a photopigment based on an

SHIFTWORK 667

opsin/vitamin A complex with peak sensitivity in the blue part of the spectrum, near 480 nm. In rodents as well as in humans this agent is more than likely OPN4 or melanopsin (Brainard et al., 2001, Thapan et al., 2001; Ruby et al., 2002; Hattar et al., 2003;).

The exploration of the human photoreceptors and the related circadian time organization requires the appropriate use and measurement of light stimuli (Foster et al., 2007). The establishment of action spectra has been helpful in associating photopigments with the responses of photobiological systems. An action spectrum for spectral sensitivity of suppression of nocturnal melatonin concentration was identified by Thapan et al. (2001) using monochromatic light exposure for 30 minutes in clinically healthy subjects. The light pulse was administered at circadian time 16–18 hours at numerous wavelengths (λmax range, 424–548 nm), and a wide range of irradiance (0.7–65.0 μW/cm2). At each wavelength, suppression of plasma melatonin increased with increasing irradiance. The action spectrum revealed a peak in sensitivity at a λmax of 459 nm, which fitted best with the rhodopsin wavelength profile. A comparable action spectrum for wavelength was obtained in humans of both genders in night time melatonin suppression tests (over a wavelength range from 420–600 nm) with 446–477 nm as the most potent region for regulating melatonin secretion (Brainard et al., 2001). The suppression of nocturnal melatonin concentrations by 1-hour light exposure of 200 or 500 lux was equal in both men and women, and proportional to the light intensity. Also, the levels and amplitude of the circadian rhythm in melatonin were not significantly different (Nathan et al., 2000). In quantifying the biological response to light, the suppression of melatonin in a constant routine protocol showed a close relation to subjective alertness, slow eye movement, and theta-α activity (detected by electroencephalography, Cajochen et al., 2000). These studies showed that light intensities as found in usual room-light illumination (90–180 lux) already have alerting and melatonin-suppressing effects (Bojkowski et al., 1987a; Cajochen et al., 2000; Zeitzer et al., 2000). In clinically healthy subjects, maximal alertness in response to light exposure was found at very short wavelengths (420, 440 and 470 nm) of the visible spectrum (Revell et al., 2006).

Human photosensitivity measured by melatonin suppression depends in part also on prior light exposure of the subjects. It increases after prior exposure of the subjects to dim light, indicating an adaptation of the photoreception or photoresponse to the recent photic history (Owen & Arendt, 1992; Hébert et al., 2002; Smith et al., 2004).

The blocking of the biologically most active short-wavelength light by the use of goggles that excluded all wavelengths of less than 530 nm prevented the suppression of the nocturnal salivary melatonin concentrations by 800 lux light intensity (Kayumov et al., 2005). All subjects (11 men and eight women, 24.7 ± 4.6 years of age) preserved their melatonin levels in filtered light similar to their dim-light secretion profile, while unfiltered bright light drastically suppressed melatonin production. Normalization of the nocturnal melatonin production by elimination of short-wavelength light apparently did not impair the measures of performance, subjective sleepiness, or alertness. Relatively good colour recognition was maintained and visual light transmittance with the filters used was approximately 73% (Kayumov et al., 2005).


Bright light exposure was able to phase shift and reset the circadian phase (Broadway et al., 1987; Czeisler et al., 1986). Also, much lesser light intensity like normal room light, (approximately 180 lux) produced a phase shift (Boivin et al., 1996), and even very dim light (20 lux) was able to synchronize the circadian system in a subject following a regular sleep, wakefulness and meal schedule (Klerman et al., 1997). In these studies with very low light intensity, the time of the daily dark span with sleep of the subjects may have reinforced the synchronizing effect of light. In a study carried without these time cues (scheduled sleep, darkness, activities), 14 days exposure of subjects to a schedule of light dark (LD) 12h:12h with 200 lux:< 8 lux were unable to maintain the initial circadian phase position (Middleton et al., 2002).

Circadian phase shift after exposure to monochromatic short-wavelength light (with two peaks at 436 and 456 nm) in a 4-hour pulse mode (8 lux, 28 μW/cm2) after habitual wake time led to a phase shift of the human melatonin profile comparable to an exposure to white light (12000 lux, 4300 μW/cm2), in spite of the white light pulse containing 185-fold more photons than the short-wavelength light (Warman et al., 2003).

Similarly, the circadian phase resetting of the free running plasma melatonin rhythm in clinically healthy subjects after a 6.5-hour exposure to monochromatic light at 460 nm induced a 2-fold greater circadian phase delay than the same time of exposure to 555 nm monochromatic light of equal photon density (Lockley et al., 2003).

The sensitivity of the circadian pacemaker varies according to the resetting effect of retinal light exposure depending upon the circadian phase at which the light exposure occurs (Honma et al., 1987; Dawson et al., 1993; Van Cauter et al., 1993). Using pre- and post-stimulus constant routines in dim light (approximately 2–7 1ux) with maintained wakefulness in a semi-recumbent posture, Khalsa et al. (2003) described a phase–response curve to a bright light exposure stimulus consisting of 6.7 hours first of a 6-minute fixed gaze exposure to 10000 lux followed by 5000–9000 lux for the remainder of the time span. Plasma melatonin was used to describe the phase of the onset, offset and midpoint of the melatonin profile. The resultant phase–response curve of the midpoint of the melatonin rhythm (with a peak-to-trough amplitude of 5 hours) showed phase delays when the light stimulus was centred before the critical phase of the core body temperature minimum, and phase advances when the stimulus was centred after the critical phase. No phase shift occurred when the stimulus was centred at the critical phase (the body core temperature minimum).

4.1.8 Photoperiod and seasonal variations

In addition to information on onset and offset of the daily photoperiod, melatonin provides information on day length. The duration of the melatonin secretion in animals and humans varies with the length of the dark span. The longer the dark span in the laboratory or the night in nature, the longer the time of melatonin synthesis and secretion, irrespective of whether the dark span is the time of activity in nocturnal rodents or of rest in diurnally active species, including humans (Arendt, 1995). Most mammals use the

SHIFTWORK 669

changes in the length of the daily light and dark period to detect a change in seasons, and to regulate seasonal behaviour and/or synchronize circannual rhythms (Tamarkin et al., 1985; Goldman, 2001). Seasonal variation in reproduction is directly controlled by the relative length of the light–dark span (Lincoln, 2002).

Under laboratory conditions imitating the winter season (short photoperiods), a longer sleep phase (recorded by electroencephalography) and a longer duration of nocturnal melatonin secretion was observed in human subjects (Wehr 1991, 2001). However, in the modern urban electrified environment, these changes are masked and not always detectable. In general, investigators found no seasonal change in duration of melatonin secretion at low- or mid-latitudes (Illnerová et al., 1985; Bojkowski & Arendt 1988; Matthews et al., 1991). In contrast, seasonal change with longer duration of melatonin secretion in winter was found at subpolar and polar high latitudes with marked changes in photoperiod and luminosity (Beck-Friis et al., 1984; Martikainen et al., 1985; Kauppila et al., 1987; Makkison and Arendt 1991; Levine et al., 1994), with higher daytime melatonin concentrations reported (Rönnberg et al., 1990). Also, when people spent more time outdoors in the summer, even in temperate climate (mid-latitude), seasonal changes in secretion of melatonin and of cortisol were found (Vondrasová et al., 1997). Kauppila et al., (1987) suggested that these elevated melatonin concentrations may be associated with diminished reproductive function. A photoperiodic influence on human fertility was observed, resulting in increased fertility in spring, but appeared to be modified by different lifestyles (Wehr, 2001).

4.1.9 Melatonin in relation to the circadian system

It is well established that ocular light exposure in humans can affect hormonal secretion, either acutely as a direct response to the presence or absence of retinal light exposure, or indirectly as a result of the influence of light on circadian mechanisms. Indeed, light is the most powerful circadian synchronizer in humans (Czeisler & Wright, 1999), and can exert a profound effect on the phase and amplitude of the human circadian pacemaker (Czeisler & Klerman, 1999). Of particular interest in the context of melatonin as a biomarker is the effect of light on the pineal function in humans: nocturnal illumination of sufficient intensity completely suppresses melatonin production (Lewy et al., 1980; Bojkowski et al., 1987a); there is considerable individual variability in sensitivity to light at night (McIntyre et al., 1990; Hébert et al., 2002; Herljevic et al., 2005); there appears to be a dose–response to light at night in that the brighter the light, the greater the reduction in nocturnal circulating melatonin (Bojkowski et al., 1987a; McIntyre et al., 1989; Zeitzer et al., 2000); bright light shifts the phase of melatonin rhythm, with morning hours being associated with phase advance and evening hours with phase delays (Duffy & Wright, 2005); and light quality during the day affects night time melatonin production (Wehr et al., 1995; Wehr, 1996; Lewy et al., 1987; McIntyre et al., 1990; Boivin et al., 1996), as well as the human circadian pacemaker (Czeisler et al., 1986).


Because melatonin is the best marker of internal clock timing and is quantifiable in the urine via well proven and reliable techniques applicable to non-laboratory studies, it has become a powerful tool as a biomarker of circadian dysregulation.

(a) Laboratory-based studies of melatonin and exposure to light at night Using sleep laboratory-based protocols, several studies have used melatonin

measurements to determine phase advance and delays resulting from controlled exposure to light at night. Deacon & Arendt (1996), Eastman and Martin (1999), Burgess et al. (2002), and Revell & Eastman (2005) have used the ‘nudging’ technique to simulate circadian rhythm disturbance in a laboratory environment. Progressively changing the timing of bright light exposure day by day (nudging) leads to a synchronized shift of the circadian system to a desired new phase. This method has been used to prepare astronauts for space flights (Eastman et al., 1995). Using bright light exposure for 9 hours on five consecutive days, the same authors reported that urinary aMT6s acrophase took at least 5 days post-treatment to return to normal baseline pattern (Deacon & Arendt, 1996). Van Cauter et al. (1994) exposed volunteers to 3 hours of bright light (5000 lux) during constant routine conditions following a 7-day entrainment period, measured plasma melatonin at 20-minute intervals, and reported rapid phase shifts within 24 hours of bright light exposure. Results after exposure for 6.5 hours to light of dim to moderate intensity early in the biological night (Zeitzer et al., 2000) showed that even small changes in ordinary light exposure during the late evening hours can significantly affect both plasma melatonin concentrations and the entrained phase of the human circadian pacemaker (Zeitzer et al., 2000). Roach et al. (2001) conducted a simulated night shiftwork study in which participants ‘worked’ during seven consecutive 8-hour shifts, and reported a mean phase delay of 5.5 hours by Night 7, using salivary melatonin measurements to detect the phase shift. A later study by the same team reported similar results using urinary aMT6s levels to assess phase delay (Roach et al., 2005).

(b) Field studies assessing the effects of shiftwork on melatonin secretion Several studies have used measurements of melatonin in the blood or urine to

evaluate and describe the effects of shiftwork on the circadian rhythm. Touitou et al. (1990) found that fast-rotating shiftwork modifies peak values and rhythm amplitudes of serum melatonin.

A series of studies on offshore oilrigs using aMT6s as a marker of internal timing and of melatonin suppression have shown complete circadian adaptation in some, but not all offshore schedules. When night-shift adaptation occurs, subjects experience desynchrony and melatonin suppression (approx. 20%) on the subsequent day shift. Thus, in addition to concerns for the health of unadapted night-shift workers, one should consider the implications of adaptation for subsequent health effects (Midwinter & Arendt, 1991; Ross et al., 1995; Gibbs et al., 2002, 2007).

SHIFTWORK 671

Using data from the Nurses Health Study II, Schernhammer et al. (2004) reported a significant inverse association between increasing number of nights worked within the 2 weeks preceding morning-void urine collection and urinary melatonin levels. Along similar lines, Hansen et al. (2006) reported lower 24-hour urinary concentrations of aMT6s in nurses working the night shift compared to nurses working the day shift; urinary concentrations of aMT6s were also lower during a day off for night-shift workers, relative to day-off levels in day-shift workers. Several authors have reported inter-individual variability in the response of the melatonin rhythm to night shift (e.g. Sack et al., 1992, 1997; Dumont et al., 2001;; Gibbs et al., 2007). In a study conducted by Quera-Salva et al. (1996, 1997) rapid change in sleep time and melatonin acrophase was reported in some night-shift workers, but not others, suggesting that some people have a physiological ability to readily adapt to rotating shift schedules, and reported for the first time a corresponding rapid shift in melatonin secretion. Dumont et al. (2001) measured urinary aMT6s every 2 hours during a 24 hour period after three consecutive nights of work in another group of nurses. Using cosinor analysis to estimate phase position, they reported individual variability in adaptation to night shiftwork, with five participants showing a phase delay, three a phase advance, and the remaining 22 demonstrating no phase shift (i.e. the timing of their melatonin secretion was typical of a day-shift worker). In a study involving offshore oil workers employed in a 1-week alternating shift schedule (1 week of nights, 1 week of days), Gibbs et al. (2002, 2007) reported adaptation to the night-shift schedule via a delay of aMT6s at the end of the week of night shifts. In another study of offshore oil workers employed in a 2-week alternating shift schedule (2 weeks of days 06:00h-18:00h, 2 weeks of nights 18:00-06:00h), Barnes et al. (1998a) reported similar results, with the delay of aMT6s occurring during the first week of the night shift. They also conducted another study in which urine samples were collected every 2 to 3 hours throughout the wake period (subjective days) and one collection over the sleep period (over-sleep sample) from a group of offshore oil workers employed in a 1-week alternating shift schedule (1 week of days (12:00-00:00h), 1 week of nights (00:00-12:00h)), and reported differing adaptation to the night shift depending on season of the year, using measurement of urinary aMT6s to detect phase shift (Barnes et al., 1998b). Prior to this, Midwinter and Arendt (1991) reported differing shifts in the acrophase of melatonin secretion depending on the season of the year, using urine samples collected over 48 hours during a week of night-shift work in a group of workers stationed in the Antarctic; they further reported a slower readaptation to the rhythm following night-shift work during the winter compared to the summer.

A study conducted by Burch et al. (2005) compared melatonin levels among workers on permanent day, swing, and night shifts. Urinary aMT6s was measured in post-work and post-sleep samples, and disrupted circadian melatonin production was evaluated using the sleep:work aMT6s ratio. They reported that night workers had altered melatonin levels, disrupted sleep, and elevated symptom prevalence. Subjects grouped by their sleep:work aMT6s ratio rather than shift had even greater symptom prevalence. Risks for two or more symptoms were 3.5 to 8 times greater among workers with sleep:work ratios


≤ 1 compared to those with ratios > 1 (Burch et al., 2005). The sleep:work ratio may be an objective means to assess circadian disruption.

(c) Melatonin as an indicator of diurnal type Several studies have used melatonin measurements to compare whether diurnal type

(morning versus evening) is associated with cumulative nocturnal melatonin secretion and/or onset of melatonin secretion. Madokoro et al. (1997) reported pronounced inter-individual differences in plasma melatonin concentration measured before and 1 year after beginning shiftwork. They constructed a ratio of melatonin concentration measured at 6 am to total melatonin concentration measured during the night, and reported that a higher Morningness-Eveningness score (indicating morning type according to Horne & Ostberg, 1976) was correlated with this ratio. Gibertini et al. (1999) measured nocturnal melatonin levels in blood on an hourly basis, and assessed the circadian type of each participant; they found that the circadian type was strongly related to the melatonin acrophase but not to the amplitude or the time of year of assessment, and that morning types experienced a more rapid decline in melatonin levels after the peak relative to evening types. Similarly, Liu et al. (2000) reported that morning type was associated with an earlier melatonin acrophase. More recently, Griefahn et al. (2002) found that the onset of melatonin synthesis was 3 hours earlier in morning than in evening types, using hourly salivary melatonin measurements; they also reported melatonin measurements to be a better indicator of diurnal type than rectal temperature measurements. Diurnal preference may be related to the ability to cope with shiftwork. Importantly, Duffy et al. (1999) and Wright et al. (2005) have established the close relationship between individual ‘free-running’ periodicity, entrained phase (entraining in a normal environment) and diurnal preference (Horne-Ostberg score).

(d) Melatonin as chronobiotic agent Exogenous melatonin given at the right biological time can synchronize (Lockley et

al., 2000; Sack et al., 2000; Arendt, 2000) and phase-shift the circadian time organization (Arendt et al., 1984; Arendt, 1985). Depending upon the stage of the circadian rhythms of a subject at a given clock hour, melatonin is able to shift circadian timing to both later and earlier times (Lewy et al., 1992). Appropriate timing of treatment to delay or advance can be predicted from a phase–response curve in subjects whose body clock phase is known (Lewy et al., 1998).

Low doses (0.3–10 mg) of melatonin given during the ‘biological day’, when endogenous melatonin levels are low, can induce sleepiness or sleep, and lower body temperature (Cagnacci et al., 1992; Deacon & Arendt, 1995, Arendt, 1995; Brzezinski et al., 2005). A single melatonin treatment (5 mg fast release) given late afternoon in controlled studies can advance the timing of the internal clock by up to about 1.5 hours (Deacon & Arendt, 1995).

SHIFTWORK 673

Timed melatonin administration (0.5–5.0 mg) given at 24-hour intervals, usually at desired bedtime can fully entrain the free-running (non-24 hour) circadian rhythm of most blind subjects (Lockley et al., 2000; Sack et al., 2000; Arendt, 2005). By acting as a circadian coupling agent countering desynchrony among central and peripheral clocks, and optimizing phase with respect to external time cues, cellular and system processes may be optimized and defence systems augmented (Arendt, 2005), with a broad range of potential therapeutic applications to be explored, including in medical oncology (Lissoni et al., 1994a,b, 2003; Bartsch & Bartsch, 2006).

4.1.10 Melatonin receptors

Melatonin displays pleiotropic physiological functions. Owing to its small lipophilic structure, it can freely enter cells and can exert an effect independent of the specific receptors found widely in human tissues (Morgan et al., 1994; Morgan & Williams, 1996; Dubocovich & Markowska 2005). For example, in over 130 locations within the brain alone (Masson-Pévet et al., 1994; Pévet et al., 2006; Wu et al., 2006) melatonin receptors are co-localized with vasopressin-, oxytocin- and corticotropin-releasing neurons. In mammalian, two subtypes of high affinity membrane receptors for melatonin have been cloned, the MT1 and MT2 subtypes (Reppert et al., 1994, 1995). They belong to the super-family of G-protein-coupled receptors. The two types show a different action spectrum with large variability among species in their distribution. The receptors are responsible for the chronobiological effects of melatonin at the level of the SCN. In most mammalian systems, MT1 receptors modulate neuronal firing, arterial vasoconstriction, affect the cell proliferation in cancer and other cells, and regulate reproductive and metabolic functions. Activation of MT2 melatonin receptors phase-shift circadian rhythms of neuronal firing in the suprachiasmatic nucleus, inhibit dopamine release in the retina, induce vasodilation and inhibit leukocyte movement in arterial beds, and enhance immune responses (Dubocovich and Markowska 2005).

Endogenous pineal melatonin is believed to feed back onto the master clock in the SCN and regulate neuronal activity and circadian rhythmicity through activation of the specific MT1 and MT2 receptors (Gillette and Mitchell 2002). The response to receptor activation varies with the circadian stage of the cell systems involved. Neurons in the SCN are most sensitive to inhibition of neuronal firing by melatonin in diurnally, as well as nocturnally, active species at dusk suggesting changes in clock excitability, possibly as expression of an endogenous circadian rhythm (Stehle et al., 1989). Melatonin phase-shifts circadian rhythms with two windows of sensitivity corresponding to the day–night (dusk) and night–day (dawn) transitions (Dubocovich et al., 1998; McArthur et al., 1997). The MT1 and MT2 receptors are targets for drug development with receptor agonists and antagonists, which are of interest for eliciting or blocking the wide variety of actions related to melatonin (Zlotos, 2005).


4.2 Proposed mechanisms for carcinogenicity of shiftwork and circadian disruption

Epidemiological studies on genetic polymorphisms in clock-related genes and phenotypes such as morning/evening preference and depressive symptoms, have shown a significant association between a single-nucleotide polymorphism in the PER3 gene and diurnal preference. In a wider sense, the circadian clock may function as a tumour suppressor at the systemic, cellular, and molecular levels. Clock-controlled genes and related factors involved in cell-cycle control include c-Myc, Mdm2, Tp53 and Gadd45a, as well as caspases, cyclins, and various transcription factors. In transgenic mice, a deletion in Per2 results in a shorter circadian period, a higher susceptibility to radiation-induced tumours, and reduced apoptosis in thymocytes. Disruption of the circadian rhythm in mice is associated with an accelerated growth of malignant tumours.

4.2.1 Melatonin and cancer

(a) Oncostatic effects of melatonin

Thirty years ago, it was hypothesized that diminished pineal function may promote the development of human breast cancer (Cohen et al., 1978). The primary argument was that increased pineal calcification, presumably leading to lowered melatonin production, was most strongly associated with increased breast cancer risk. Although this was the first reference to environmental lighting, which necessarily includes both sunlight and artificial light, as a potential source of one of several endocrine abnormalities that may underlie the development of breast cancer, light at night was not specifically postulated as an etiological factor. It was proposed incorrectly that altered visual stimulation, by blindness or darkness, would impair pineal melatonin production, thereby leading to unopposed estrogen secretion and increased breast cancer risk. It is now known that overall melatonin production is not compromised in blind individuals (Lockley et al., 1997) and that breast cancer risk is actually diminished in blind women (Hahn, 1991; Verkasalo et al., 1999). It was postulated by Stevens (1987) that light exposure at night may represent a unique risk factor for breast cancer in westernized societies via its ability to suppress nocturnal melatonin production by the pineal gland. This postulate, referred to as the ‘melatonin hypothesis’, was based on in-vivo studies demonstrating that melatonin inhibits, while pinealectomy or constant bright light stimulates, the development and growth of experimental breast cancer in rodents, and by in-vitro studies showing that the proliferation of estrogen receptor positive (ER+) human breast cancer cells was directly suppressed by nocturnal physiological levels of melatonin (Stevens, 2006).

Many studies using pharmacological concentrations of melatonin have demonstrated a direct antiproliferative and/or apoptotic effect on cancer cells (usually human cancer cell lines) in vitro. A substantial number of investigations have also shown that nocturnal physiological concentrations of melatonin exert direct oncostatic effects on cancer cell

SHIFTWORK 675

proliferation as well. However, several studies have also demonstrated cytotoxic effects of pharmacological levels of melatonin on cancer cells in vitro. A large number of studies in experimental animal models of tumorigenesis have shown that properly timed (i.e. relative to the light–dark cycle) administration of pharmacological doses of melatonin can inhibit the development and/or growth of a wide variety of murine tumours (i.e. chemically induced, transplantable, spontaneous) and human cancer xenografts. The mechanisms by which these tumour inhibitory effects are exerted are not totally clear but in some cases may involve the inhibition of tumour linoleic acid uptake and metabolism as well as direct oncostatic, immune enhancing, and/or free-radical/antioxidant actions (Blask, 1993, 2001; Blask et al., 2002, 2005a).

In most of these in-vitro studies, ER+ MCF-7 human breast cancer cells were cultured for several days in the presence or absence of melatonin, usually at a high physiological concentration of 1 nM. Depending on the study, the robustness of melatonin’s oncostatic effects could be quite variable ranging from 80% to less than 20% inhibition of cell proliferation. In several investigations involving either human MCF-7 human breast cancer, neuroblastoma, uveal melanoma or murine colon carcinoma cells, the dose–response to melatonin exhibited a bell-shaped pattern with the most robust antiproliferative effects occurring in nocturnal physiological range (Blask et al., 2002). In a small number of studies, no oncostatic effects of melatonin were reported in the physiological range on MCF-7 cells or on several other human cancer cell lines (human cervical carcinoma [Hela], osteosarcoma [MG-63] and lymphoblastoid [TK6]) while cytotoxic effects were observed at pharmacological levels (Panzer et al., 1998; Baldwin & Barrett, 1998; Baldwin et al., 1998). These discrepancies were most likely due to the use of different culture conditions as well as subclones of cells with lower sensitivity to melatonin (Bartsch & Bartsch, 1981; Bartsch et al., 2000).

The oncostatic action of melatonin at physiological concentrations, particularly on ER+ MCF-7 human breast cancer cells in vitro, encompasses a variety of molecular and cellular mechanisms, some of which involve the inhibition of the mitogenic action of hormones and growth factors, most notably estradiol (E2), epidermal growth factor, and prolactin. Major effects of physiological concentrations of melatonin on the cellular biology and cell-cycle control of ER+ MCF-7 cells include a slowing of the progression of cells from the G0-G1 phase of the cell cycle to the S phase (DNA synthetic phase) with a resultant lengthening of the transit time through the cell cycle. Evidence indicates that this is accomplished by a melatonin-induced increase in the expression of the tumour suppressor gene TP53 which in turn activates p21WAF1 protein expression leading to eventual cell-cycle arrest via inhibition of the ability of cyclin-dependent kinases to phosphorylate the retinoblastoma protein (Rb). Additionally, melatonin inhibits DNA synthesis in that reduced proportion of MCF-7 cells that progress to the S phase of the cell cycle (Sánchez-Barceló et al., 2003). While evidence indicates that pharmacological concentrations of melatonin can induce apoptosis in cancer cells, there is no convincing experimental evidence that programmed cell death is activated at physiological levels of this indoleamine (Cos et al., 2002).


In addition to its oncostatic effects, melatonin is able to reduce the invasive and metastatic properties of MCF-7 cells in vitro. This appears to be mediated, in part, by a melatonin-induced upregulation in the expression of cell surface proteins E-cadherin and β1-integrin (Cos et al., 1998). Melatonin at physiological levels can increase gap-junction-mediated intercellular communication between MCF-7 cells in culture (Cos & Fernández, 2000) and cause alterations in the cytoskelelal arrangements of ER+ T-47D human breast cancer cells in culture (Matsui & Machado-Santelli, 1997).

Melatonin exerts a major role as an antiestrogen in ER+ human breast cancer cell proliferation by suppressing the activity of the estrogen growth response pathway (Hill et al., 1992). In MCF-7 cells in particular, the molecular mechanisms of this antiestrogen action centre around the fact that melatonin downregulates the transcription of ERα, prevents estrogen-dependent transcriptional activation by destabilizing the E2-ER complex from binding to the estrogen-responsive element of DNA via antagonism of calmodulin interactions with ERα, and by blocking E2-induced upregulation of cyclin D1 expression (Molis et al., 1994; Rato et al., 1999; del Río et al., 2004; Cini et al., 2005). Melatonin does not bind to ERα nor does it interfere with the binding of E2 to the ligand-binding domain of ERα. These molecular events mediating melatonin’s oncostatic actions through suppression of the activity of the estrogen-response pathway appear to involve the MT1 melatonin receptor suppression of cAMP production as well as calmodulin antagonism (Ram et al., 2002; Kiefer et al., 2002; Rato et al., 1999).

(b) Melatonin, aromatase and telomerase

In-vitro and in-vivo studies have examined the influence of melatonin, either at nocturnal physiological circulating concentrations or pharmacological doses, on the local biosynthesis of estrogens from androgens via modulation of aromatase activity by ER+ MCF-7 human breast cancer cells (Cos et al., 2005) or dimethylbenzanthracene (DMBA)-induced rat mammary cancers (Cos et al., 2006), and the impact on cell proliferation and tumour growth. Melatonin downregulated aromatase expression at the transcriptional level in MCF-7 cells and reduced aromatase activity in both MCF-7 cells and DMBA-induced rat mammary tumours resulting in a diminished rate of tumour cell proliferation and growth. This anti-aromatase action of melatonin is mediated via the MT1 melatonin receptor (González et al., 2007).

Melatonin at physiological nocturnal circulating levels and pharmacological concentrations inhibits both the expression and activity of telomerase in MCF-7 human breast cancer cells in culture, and in xenografts in female nude mice. Telomerase, an enzyme responsible for the elongation of telomeres at the ends of chromosomes, is activated in most human cancers. Melatonin appears to regulate telomerase mRNA expression via both membrane and nuclear melatonin-receptor-mediated mechanisms (Leon-Blanco et al., 2003, 2004).

SHIFTWORK 677

(c) Effects of melatonin on sex hormones

In animals, the effects of physiological levels of melatonin, either endogenously produced or exogenously administered, on reproductive hormones (pituitary gonadotrophins, prolactin, gonadal steroids) are well known, particularly in animal species that breed seasonally (Goldman, 1999, 2001). In these animals, melatonin can exert either inhibitory, stimulatory or modulatory effects on these hormones depending upon the species and situation.

In humans, the situation is more problematic. While low pharmacological doses of melatonin administered to human subjects over the course of several days to a few weeks had no effect on either pituitary or gonadal hormones (Arendt, 1985; Wright et al., 1986; Luboshitzky et al., 2000; Arendt, 1995), extremely high doses of melatonin produced a slight reduction in blood levels of pituitary gonadotrophic hormones (Voordouw et al., 1992). As part of a contraceptive study, the administration of large oral doses of melatonin (300 mg) on a daily basis for 30 days to young adult women with normal menstrual cycles, caused significantly decreased mean circulating levels of luteinizing hormone (LH), estradiol and progesterone compared with non-treated controls, possibly through mechanisms dependent on (i.e. enhanced) or independent of steroid negative feedback (Voordouw et al., 1992). On the other hand, much lower doses of oral melatonin (3 mg) enhanced LH and follicle-stimulating hormone (FSH) levels in response to a gonadotropin-releasing hormone (GnRH) challenge during the follicular but not luteal phase of the menstrual cycle (Cagnacci et al., 1995a). Pharmacological levels of melatonin (2–5 mg) have been demonstrated to stimulate prolactin secretion into the blood following oral administration in adult men and women (Terzolo et al., 1993; Kostoglou-Athanassiou et al., 1998) whereas in another study, it had no effect at this dose range (Terzolo et al., 1990).

(d) Melatonin, free radical scavenging and antioxidation

The role of free radicals in oncogenesis encompasses the initiation, promotion and progression stages of tumour development and growth. For example, the exposure of normal somatic cells to chemical carcinogens can generate free radicals that can cause DNA damage that, in turn, may lead to the initiation of cancer. DNA mutations caused by free radicals may become fixed as a consequence of a wave of clonal expansion due to the relative growth advantage that new mutations confer on cells. Additional genomic instability induced by faulty cell division or defective DNA-repair mechanisms may further increase the rate of tumorigenic mutations. Moreover, free radicals and other reactive oxygen species may provide additional stimulation of signal transduction mechanisms that may lead to enhancement of cell proliferative and survival mechanisms (Marte, 2004). While there is no question that melatonin at pharmacological concentrations has potent direct free radical scavenging effects, the role of physiological levels in free radical scavenging remains controversial (Reiter et al., 2001). At


pharmacological levels, melatonin pretreatment substantially reduces the formation of DNA damage in liver tissue in rats caused by the chemical carcinogen safrole, implying that these levels of melatonin have the potential to prevent the initiation of hepatic carcinogenesisby reducing free radical-induced DNA damage. In pinealectomized rats, a further increase in DNA-adduct formation over pineal-intact controls was observed, indicating that the endogenous physiological melatonin signal confers a degree of partial protection against free-radical-induced nuclear DNA damage, and in doing so, may reduce the probability of cancer initiation (Reiter, 2001). Physiologically relevant melatonin levels have been reported to upregulate endogenous antioxidant enzyme systems such as glutathione (GSH) peroxidase, γ-glutamylcysteine-synthetase, the rate-limiting enzyme responsible for GSH synthesis as well as GSH levels, and superoxide dismutase (Blask et al., 1997; Hardeland, 2005). In the case of the enzyme for GSH synthesis, this upregulation appears to be mediated via a mechanism mediated by the melatonin receptor (Hardeland, 2005). Not only do physiological concentrations of melatonin markedly elevate total GSH concentrations in MCF-7 human breast cancer cells, but adequate intracelluar levels of GSH appear to be an absolute requirement for the oncostatic action of melatonin in these cells in vitro. Furthermore, ER– HS578T human breast cancer cells that are ordinarily insensitive to the oncostatic action of physiological melatonin concentrations can be coerced into responding to melatonin by raising intracellular GSH levels (Blask et al., 1997). Paradoxically, physiological levels of melatonin have been shown to actually enhance the production and release of reactive oxygen species into the incubation medium by human monocytes co-cultured with human cancer cell lines. This resulted in increased lethality to the cancer cells, indicating a beneficial pro-oxidant effect for melatonin (Morrey et al., 1994).

4.2.2 Circadian genes and cancer: possible mechanisms

(a) Genetic determinants of circadian rhythms Although the daily oscillation of physiological and behavioural processes in plants

and animals were observed thousands of years ago, it wasn't until the 1960s that such oscillating rhythms were found to be regulated genetically (Pittendrigh, 1967). The first circadian gene, Period, was cloned from fruitflies in the mid-1980s (Bargiello et al., 1984; Reddy et al., 1984). Since then, rapid advances in the field of circadian biology have revealed that these clocks are operated by numerous gene products that function in interacting feedback loops in all species studied. Biological clocks provide organisms with a survival advantage, by organizing their behaviour and physiology around cyclic changes in the environment.

At the time of writing, nine core circadian genes have been identified: Clock (King et al., 1997), casein kinase I epsilon (CSNK1E) (Takano et al., 2000), cryptochrome 1 (CRY1), cryptochrome 2 (CRY2) (Hsu et al., 1996), Period 1 (PER1), Period 2 (PER2), Period 3 (PER3) (Shearman et al., 1997; Tei et al., 1997), neuronal PAS domain protein 2

SHIFTWORK 679

(NPAS2) (Reick et al., 2001), and aryl hydrocarbon receptor nuclear translocator-like (ARNTL) (also referred to as brain and muscle Arnt-like protein-1, BMAL1) (Hogenesch et al., 1998; Honma et al., 1998). The three PER genes encode PER–ARNT–Single-minded protein (PAS)-domain proteins that function as surfaces allowing heterodimerization among different clock proteins. The CLOCK and BMAL1 genes encode basic–helix–loop–helix (bHLH)-PAS transcription factors. NPAS2 is a paralogue of the transcription factor CLOCK, which is a major player in the SCN. The human CRY genes encode proteins similar to plant blue-light receptors within class I photolyases. A common feature of these circadian genes is that the levels of mRNAs and proteins that they code for, with the exception of those coded for by CLOCK and CSNK1E, oscillate throughout a 24-hour period (Reppert & Weaver 2001).

(b) Circadian genes and clock control The circadian system is divided into two parts, the central pacemaker and the

peripheral clocks. The mammalian circadian clock contains three components: input pathways, the central pacemaker, and output pathways. The input pathways transmit information from environmental cues to the central pacemaker. The central pacemaker synchronizes with the environment to generate endogenous rhythms. The output pathways convert the instructions from the central pacemaker into daily oscillations in various physiological and behavioural processes (Hastings, 2000; Hastings & Herzog, 2004; Fu & Lee, 2003). The central pacemaker in mammals resides in the SCN of the anterior hypothalamus. The SCN is composed of multiple, single-cell circadian oscillators that can generate coordinated circadian outputs when synchronized (Welsh et al., 1995; Liu et al., 1997).

A model of transcription-translation feedback loops of circadian genes has been proposed to explain the molecular clockwork of the mammalian central pacemaker (Reppert & Weaver, 2001, 2002; Young & Kay, 2001). At the molecular level, the circadian clock is organized as positive and negative feedback loops based on transcription-translation. The positive components of the loops are CLOCK (or NPAS2) and BMAL1, which form a heterodimer that regulates the expression of genes containing E-box regulatory segments in their promoter regions (Reppert & Weaver 2001; Fu et al., 2002). This heterodimer directly induces the genes coding for the PERs and CRYs, which constitute the major components of the negative feedback loop. An additional level of regulation within the positive feedback loop is provided by REV-ERBα, which controls cyclic BMAL1 expression (Preitner et al., 2002). The CLOCK–(or NPAS2–)BMAL1 complex also regulates the expression of multiple other clock-controlled genes, either directly or indirectly.

Similar interacting loops of core circadian gene products regulate circadian rhythms in peripheral tissues. The “peripheral clocks” are regulated by the SCN pacemaker, through both the autonomic nervous system and neuroendocrine systems (Bartness et al., 2001; Kalsbeek & Buijs 2002). The rhythmic expression of core circadian genes is observed in most peripheral tissues (Zylka et al., 1998), and can be induced in cultured


fibroblasts (Balsalobre et al., 1998). Ablation of the SCN has been shown to abolish circadian gene oscillation in peripheral tissues as well (Balsalobre et al., 1998; Sakamoto et al., 1998). These findings suggest that the peripheral clocks are either driven or synchronized by the SCN pacemaker. Both the SCN central clock and peripheral tissue clocks regulate cell functions by affecting the expression of clock-controlled genes. Studies have indicated that 2–10% of all mammalian genes are clock-controlled genes (Le Minh et al., 2001; Duffield et al., 2002; Storch et al., 2002).

Recent data also reveals that microRNAs play a key role as regulators of the circadian-timing process. miR-219–1 is a clock-controlled gene that plays a role in regulating the length of the circadian rhythm. miR-132 is light-inducible and modulates the phase-shifting capacity of light. Both microRNAs potentially regulate both clock periodicity and clock entrainment (Cheng et al., 2007).

Time information from the central oscillator to the peripheral oscillators is transmitted by neural and humoral stimuli (Schibler et al., 2003; Buijs & Kalsbeek, 2001), which are currently not well characterized but among which a secretory product of the pineal gland, melatonin, is thought to play a prominent role. The central oscillator in turn is kept in step (“synchronized”) with the periodic surrounding by the light–dark alteration of the astronomical calendar day. The oscillators in the peripheral tissues are also subject to the entraining (synchronizing) effects of social routine, physical exercise, and food uptake. These secondary synchronizers determining the phase setting of many peripheral oscillators may compete with the lighting regimen, acting over the central oscillator in the SCN in a multifactorial way (Challet & Pévet, 2003). In some tissues, like the liver, the time of food uptake may become the dominant synchronizer and determine the staging of the metabolic rhythms in this organ (Stokkan et al., 2001; Hara et al., 2001). In daily life, circadian rhythms determine the rhythmically varying degree of cognitive functioning and physical strength and dexterity, resulting in predictable timing of best and worst work performance and efficiency. Optimal function of the human body requires a certain sequential and phase-related ordering of the many circadian rhythms extending from the molecular to organismic level.

(c) Circadian gene polymorphisms and circadian disturbance in humans A limited number of epidemiological studies have examined genetic polymorphisms

in clock-related genes and phenotypes such as morning/evening preference and depressive symptoms (Johansson et al., 2003). The T3111C variation of the CLOCK gene was first reported by Katzenberg et al. (1998) to be associated with morning/evening preference as assessed by the Horne-Ostberg scale (Horne & Ostberg, 1976). A single nucleotide polymorphism located in the 3′ flanking region of the human CLOCK gene was demonstrated to be a predictor of diurnal preference in a population-based random sample of 410 normal adults. A smaller study of 105 subjects, however, did not confirm this association (Robilliard et al., 2002). In addition, associations of PER3 polymorphisms with delayed sleep phase syndrome or diurnal preference have recently been reported (Ebisawa et al., 2001; Archer et al., 2003; Johansson et al., 2003). Archer

SHIFTWORK 681

et al. (2003) used the Horne-Ostberg scale to examine a PER3 length polymorphism in which the longer allele was associated with ‘morningness’ and the shorter allele with ‘eveningness’. The shorter allele was also strongly related to delayed sleep phase syndrome, and they reported allele frequencies to be 68% for the shorter allele and 32% for the longer allele. A recent study found that the homozygous PER3 longer allele (5/5) had a considerable effect on sleep structure and waking performance (Viola et al., 2007). Rapid eye movement (REM) sleep was increased in PER3 (5/5) compared to PER3 (4/4) individuals. In addition, the decrement of cognitive performance in response to sleep loss was significantly greater in the PER3 (5/5) individuals. By contrast, the circadian rhythms of melatonin, cortisol, and peripheral PER3 mRNA expression were not affected. These findings show that this polymorphism in PER3 predicts individual differences in the sleep-loss-induced decrement in performance, and the differential susceptibility may be mediated by its effects on sleep homeostasis. Johansson et al. (2003) examined a single nucleotide polymorphism in the PER3 gene (647 Val/Gly) in a Swedish population and their results showed a significant association between this PER3 genetic variation and diurnal preference.

(d) Circadian genes: potential tumour suppressors A novel role of circadian genes in tumorigenesis comes from discoveries

demonstrating that the circadian clock may function as a tumour suppressor at the systemic, cellular and molecular levels through its involvement in cell proliferation, apoptosis, cell cycle control, and DNA-damage response.

(i) Regulation of cell cycle and apoptosis The cell cycle is regulated by an internal circadian clock. In cells of peripheral tissues,

the circadian clock controls cell proliferation and apoptosis by regulating the expression of circadian-controlled genes. The circadian clock mechanism is directly involved in the regulation of cell division (Reddy et al., 2005; Lee, 2005; Gery et al., 2006). Studies have shown that about 7% of all clock-controlled genes identified in rodents regulate either cell proliferation or apoptosis (Kornmann et al., 2001; Akhtar et al., 2002; Duffield et al., 2002; Panda et al., 2002). These clock-controlled genes include c-Myc and Mdm2, the tumour-suppressor genes Tp53 and Gadd45a, as well as genes that encode the caspases, cyclins, transcription factors, and ubiquitin-associated factors that are involved in regulating the cell cycle and apoptosis. In humans, the rhythmic expression of several cyclins and the tumour-suppressor p53 are also regulated by the circadian clock (Bjarnason et al., 1999). Moreover, the expression patterns of these clock-controlled genes are synchronized with the circadian oscillation patterns of PER1 and BMAL1 expression in the same tissue (Bjarnason et al., 2001). The Per2 gene may also play an important role in tumour suppression by inducing apoptotic cell death by enhanced pre-apoptotic signalling and attenuated anti-apoptotic processes (Hua et al., 2006).


(ii) Modulation of cell proliferation In addition to controlling the expression of cell-cycle genes and tumour-suppressor

genes at the transcriptional and post-transcriptional levels, the core circadian genes are also involved directly in modulating the intracellular signalling pathways that regulate cell proliferation. It has been shown that a core circadian regulator, CSNK1E also functions in promoting cell proliferation by stabilizing β-catenin (Lee et al., 2001). β-Catenin can interact with transcription factors of the T-cell-specific transcription factor/lymphoid enhancer factor-1 family to regulate transcription (van de Wetering et al., 2002), and promote tumorigenesis (Morin, 1999).

(iii) Control of the cell-cycle checkpoint Whether the circadian clock and the cell-cycle clock are connected in vivo was

subject to debate until a recent study of a mouse model demonstrated that apoptosis induced by γ-radiation is dependent on circadian time in both wild-type and Per2-mutant thymocytes (Fu et al., 2002). Specifically, when irradiated at the early stage of the active phase or at the early stage of resting phase, Per2-mutant thymocytes show a G2/M-specific resistance to radiation-induced apoptosis. Therefore, the circadian clock not only regulates the expression of cell-cycle genes but could also be involved in controlling cell-cycle checkpoint function.

(iv) Response to DNA damage Studies from animal models have also shown that the core circadian genes can

respond directly to γ-radiation. However, disruption of the Per2 gene stops the response of all core circadian genes to γ-radiation (Fu et al., 2002), suggesting that the molecular clock itself can be modulated by genotoxic stress in peripheral tissues. The ability of circadian genes to mediate the DNA-damage response seems to be cell autonomous, since Per2-mutant thymocytes have been shown to attenuate p53-induction in response to γ-radiation in vitro. Per2-mutant mice also show altered cell division, increased sensitivity to ionizing radiation with impaired DNA repair, and development of malignancies (Fu et al., 2002; Matsuo et al., 2003). Besides γ-radiation, it has also been shown that the clock genes can respond to low levels of ultraviolet irradiation in cultured human keratinocytes (Kawara et al., 2002). Per1 also plays an important role in regulating growth and DNA-damage control, and it interacts with proteins in the cell-cycle pathway (Gery et al., 2006).

The findings that circadian genes are involved in cancer-related biological pathways such as cell-cycle control and DNA repair support the assumption that disturbances in circadian oscillator functions may increase the risk of carcinogenesis in a variety of tissues. Animal experiments also suggest the genetic circadian oscillator system may be involved in carcinogenesis. These reports have justifiably raised widespread health concerns.

SHIFTWORK 683

(e) Loss of circadian gene functions in tumorigenesis In animal models, mice with disruptions in the core circadian gene Per2 have recently

been shown to display salivary-gland hyperplasia and develop spontaneous lymphoma (Fu et al., 2002). It is likely that deregulation of multiple molecular pathways contribute to the cancer-prone phenotype of the Per2-mutant mice. Deregulation of the Myc-mediated growth-regulatory pathway is proposed to be one possible mechanism by which disruption of the circadian clock could promote tumour formation. Zheng et al. (1999) constructed a mouse in which there was a deletion mutation in Per2, resulting in a shorter circadian period (~22 hours). Fu et al. (2002) reported that this mouse was also more susceptible to radiation-induced tumours, and showed reduced apoptosis in thymocytes. This mouse showed temporal deregulation of cyclin D1, cyclin A, and c-Myc. Moreover, the disruption of circadian rhythms in mice is associated with accelerated growth of malignant tumours, suggesting that the host circadian clock may play an important role in endogenous control of tumour progression (Filipski et al., 2002).

In humans, direct evidence has demonstrated an association between the loss of human PERIOD (hPER1 and hPER2 and hPER3) function and human breast and human endometrial cancer. One study showed that ~95% of breast tumours (53 out of 55 specimens) displayed no or deregulated levels of PER1, PER2 or PER3 proteins in the breast tumour cells when compared to the adjacent normal tissue (Chen et al., 2005). In another study, it was also observed that the loss of PER1 protein was common in human endometrial carcinoma but not in the adjacent normal cells (Yeh et al., 2005). Chen et al. (2005) further suggested that the loss of clock-gene expression was linked to DNA-methylation of clock-gene promoters rather than genetic mutations of the clock genes (see paragraph below). A recent study also showed a downregulation of PER1 in human breast and lung cancer tissue (Gery et al., 2006). It was suggested that altered PER expression, resulting in the disruption of normal circadian clock control, may benefit the survival of cancer cells and promote carcinogenesis.

(f) Epigenetic alterations of circadian genes in cancer tissue The expression level of a gene can be dramatically influenced by the methylation

status of its promoter region, and alterations in global methylation patterns have been associated with cancer development. Similarly, changes in circadian gene methylation patterns have also been observed in cancer tissue. One study revealed disturbances in the expression of the three period genes in over 95% of breast cancer cells when compared to non-cancerous cells (Chen et al., 2005). Promoter methylation of PER1, PER2, and/or CRY1 was detected in one-third of endometrial cancers compared to one-fifth of non-cancerous endometrial tissues (Shih et al., 2006). The expression levels of PER1, PER2, PER3, CRY1, CRY2 and BMAL1 were also significantly impaired for those having chronic myeloid leukaemia, and the promoter region of PER3 was found to be methylated in all of these leukaemia patients (Yang et al., 2006). Microarray analysis also identified PER1 as being expressed at lower levels in non-small cell lung cancer tissue compared to


normal tissue, and artificially induced expression of PER1 in non-small cell lung cancer cell lines resulted in a significant reduction in growth. It is believed that DNA hypermethylation and histone H3 acetylation are potential mechanisms for this silencing of PER1 expression (Gery et al., 2007).

(g) Evidence from genetic association studies in humans The human PER3 gene was the first circadian gene to be linked to increased risk of

breast cancer in a genetic association study (Zhu et al., 2005). The PER3 gene, which is a central component of the clockwork mechanism, contains a polymorphic repeat region comprised of four or five copies of a 54 bp repetitive sequence in exon 18. This length variation results in the insertion/deletion of 18 amino acids and has been found to be associated with delayed sleep phase syndrome and diurnal preference (Ebisawa et al., 2001; Archer et al., 2003). A case–control study found that the variant genotypes (heterozygous + homozygous 5-repeat alleles) were associated with an increased risk of breast cancer among premenopausal women (OR, 1.70; 95% CI: 1.00–3.0) (Zhu et al., 2005).

NPAS2, the largest circadian gene, is a member of the basic helix–loop–helix-PAS class of transcription factors. NPAS2 forms heterodimers with BMAL1, which transcriptionally activates PER and CRY expression, which are required for maintaining biological rhythms in many organisms (Vitaterna et al., 1994; Bunger et al., 2000). The NPAS2 gene is very conserved, with only one missense mutation (SNP database accession No. rs2305160, Ala394Thr located in alternative exon 22), listed in the NCBI SNP database. A recent breast cancer case–control study found that women with the heterozygous Ala394Thr genotype had a significantly reduced breast cancer risk compared to those with homozygous Ala394Ala (OR, 0.61; 95%CI: 0.46–0.81) (Zhu et al., 2008). Furthermore, this protective role was more evident in premenopausal women (OR, 0.44; 95%CI: 0.27–0.77) than in postmenopausal women (OR, 0.65; 95% CI: 0.49–0.91). This is the first piece of evidence demonstrating an association between NPAS2 and human breast cancer.

The same polymorphism (rs2305160, Ala394Thr) in the NPAS2 gene has also been genotyped in another population-based case–control study of non-Hodgkin lymphoma (Zhu et al., 2007). These results demonstrated that risk of non-Hodgkin lymphoma was significantly reduced among individuals with the heterozygous Ala/Thr genotype (OR, 0.69; 95%CI: 0.53–0.90), the homozygous variant Thr/Thr genotype (OR, 0.55; 95%CI: 0.36–0.85), and both variant Thr genotypes combined (Ala/Thr & Thr/Thr) (OR, 0.66; 95%CI: 0.51–0.85), when compared to those with the homozygous Ala/Ala genotype. Similar reduced risks were detected for B-cell lymphoma and its two major subtypes: B-cell chronic lymphocytic leukaemia/prolymphocytic leukaemia/small lymphocytic lymphoma, and follicular lymphoma. Previous studies have shown that disruption of circadian rhythm may cause disordered immune responses such as aberrant immune cell trafficking and abnormal cell proliferation cycles (Mormont & Lévi, 1997; Vgontzas & Chrousos, 2002). Given the established association between immune

SHIFTWORK 685

dysregulation and non-Hodgkin lymphoma (Filipovich et al., 1992), the observed role of the circadian genes in lymphomagenesis could be explained by their impacts on immune activity.

(h) Expression of circadian genes Mammalian circadian oscillators were originally believed to exist only in neurons of

the SCN (Ralph et al., 1990). However, with the identification of the mammalian clock and clock-controlled genes, this view has changed dramatically. Circadian oscillators have been uncovered in both central and peripheral tissues, with the SCN presumed to coordinate cyclic gene expression in the periphery by neural and/or humoral signals (Balsalobre et al., 2000; Kramer et al., 2001; Le Minh et al., 2001; McNamara et al., 2001; Cheng et al., 2002). Robust daily oscillations in gene expression could be detected in almost all investigated tissues (Schibler et al., 2003). These daily cycles were believed to be the result of cyclic humoral or neuronal signalling from the SCN. However, the autonomy of peripheral oscillators is now under debate as peripheral tissues explanted and maintained in culture demonstrate continued oscillation of Per2 for up to 20 days, and SCN lesioning does not abolish this circadian oscillation (Yoo et al., 2004). Moreover, the mechanisms of regulation of peripheral clocks and indeed, their function, remain largely obscure. Furthermore, disturbances in the expression of the three PER genes (PER1, PER2, and PER3) have been detected in human breast cancer cells in comparison with nearby non-cancerous cells (Chen et al., 2005). Because the circadian clock controls expression of cell-cycle-related genes, it was suggested that altered PER gene expression may result in the disruption of the control of the normal circadian clock, hence benefit the survival of cancer cells, and promote carcinogenesis.

(i) Light exposure and circadian gene expression Light is the most powerful circadian synchronizer among all environmental cues

(Lucas et al., 2001; Wright & Czeisler, 2002). The molecular mechanisms involved in synchronization to light have been demonstrated in previous experiments. For example, both Per1 and Per2 could be induced in SCN tissue by light exposure in mice (Albrecht et al., 1997). In the SCN of wild-type mice, light exposure also evoked a transient interaction between Protein Kinase C α and Per2 proteins that affects Per2 stability and nucleocytoplasmic distribution (Jakubcakova et al., 2007). Using oral mucosa samples, a recent study showed that the induction of human PER2 expression was stimulated by exposing subjects to 2 hours of light in the evening (Cajochen et al., 2006). The increase in PER2 expression relative to a non-light control condition was statistically significant after exposure to light at 460 nm (blue), but not after exposure to light at 550 nm (green). The authors concluded that the non-image-forming visual system is involved in human circadian gene expression (Cajochen et al., 2006).


(j) The human time structure and its alteration by phase shift The temporal organization of the human body has to be understood to appreciate the

impact of night work and shiftwork upon worker health and well-being. The human body has not only a structure in space, as expressed by its gross and microscopic anatomy, but it has a structure in time, as expressed by rhythms of numerous frequencies superimposed upon linear trends associated with development and aging (Touitou & Haus, 1992). The rhythmic variations encountered vary in period from milliseconds, e.g. in individual nerve cells, to minutes or hours (ultradian rhythms) to 24 hours (circadian rhythms), and to longer periods, such as the menstrual cycle in women, and yearly cycles (circannual rhythms) in both men and women (Haus & Touitou, 1992; Hildebrandt et al., 1998).

Rhythms of a person, synchronized to diurnal activity by the ambient light–dark cycle and social routine, must undergo phase readjustment when forced to adhere to a new activity–sleep schedule due, for example, to night work or geographic displacement across several time zones. The central and peripheral oscillators will follow the new schedule, not immediately however, but over a certain number of transient cycles, to adapt to the changed phase of the environmental synchronizer. During this time of adaptation, disruption of the usual sequence and biological order of the numerous rhythmic events takes place with some clock genes responding faster than others. The result is an internal phase desynchronization within the oscillator mechanism (Sakamoto & Ishida, 2000; Nagano et al., 2003; Nakamura et al., 2005). The circadian oscillators in the anterior region of the SCN undergo a faster time adaptation than those in the posterior portion (Nagano et al., 2003; Nakamura et al., 2005). During the re-adjustment period, desynchronization occurs within the oscillators as well as among different oscillatory tissues and brain regions that re-adjust their phases at different rates (Stokkan et al., 2001; Abe et al., 2002; Nagano et al., 2003). Within the oscillators, after a shift in the light–dark regimen, there is a faster shift of Per1 and Per2 oscillator genes and a slower shift of Cry1, another component of the oscillator mechanism (Reddy et al., 2002). In the molecular oscillator mechanism, as in the organism as a whole, there is a difference in response with different directions of the phase shift. Phase advances (earlier timing) of the lighting schedule lead to a more prolonged desynchronization within the SCN than do phase delays (Nakamura et al., 2005). Also Per1 and Per2 genes have been found to behave differently during advancing and delaying phase shifts (Yan & Silver, 2002; Albrecht et al., 1997). Moreover, the phase-shifting kinetics of circadian rhythms in transcriptional activity show region-specific differences (Nagano et al., 2003; Nakamura et al., 2005), with different tissues exhibiting different resetting behaviour than the SCN or behavioural rhythms (Abe et al., 2002). The adaptation of the peripheral oscillators is independent – in part – of the hypothalamic control (Stokkan et al., 2001). Thus, during the phase-resetting process, internal desynchronization is manifested within the individual oscillators and simultaneously also between central and peripheral oscillators. In the absence of hypothalamic control and synchronization, peripheral oscillators of diverse tissues cycle with different periods; thus, during the process of adaptation, they express

SHIFTWORK 687

different phases and changed phase relations. The unique circadian phase and period values expressed by each tissue suggest that the quantitative properties of the circadian oscillators in each tissue are unique and tissue-specific (Yoo et al., 2004) and/or may be the expression of different synchronizing mechanisms acting upon different tissue oscillators (Lakatua et al., 1975, 1983, 1988). The overall effect of a phase shift of this nature is alteration at several hierarchical levels of the internal time organization during the transitional duration of adjustment. For example, the top physical efficiency that is typically observed in the afternoon becomes delayed into the night time. The propensity to sleep, which is the expression also of a circadian rhythm, may be high during the environmental time that requires alertness and efficiency, and it may be low during the time reserved for rest, resulting in insomnia and non-restorative sleep. During adaptation, this external and internal desynchronization of the human organism leads to a functional disturbance of the time organization (“dyschronism”), with loss in performance efficiency plus the expression of a set of symptoms, similar to those of jet lag (Hildebrandt et al., 1974; Harris, 1977; Ribak et al., 1983; Folkard & Akerstedt, 2004; Folkard & Lombardi, 2004).

In this context, it is important to understand that a circadian phase shift: (1) affects all metabolizing and proliferating cells in the organism; (2) leads to transient internal desynchronization on a molecular basis within the individual cellular oscillators; (3) results in desynchronization among the cellular oscillators in the SCN and peripheral tissues; (4) is not immediate but requires time (days) for complete adjustment, occurring over several transient cycles; and (5) varies by variable and function in the amount of time required for phase adaptation and, with regard to cell and tissue proliferation, may extend over several weeks’ time.

A circadian phase shift exerts its effects upon molecular cell and tissue physiology and occurs over an extended period during which the time sequence of the biological rhythms of many variables is different from that found in day–night-adapted individuals, i.e. the circadian time organization which is thought to be linked to optimal function (Touitou & Haus, 1992, Winget et al., 1992; Monk, 1992; Mormont & Waterhouse, 2002). Changes in the neuroendocrine web, controlling cell and tissue proliferation during the internally desynchronized span of phase adaptation, may permit or even promote growth of abnormal cell proliferation in target tissues that find themselves out of phase with their usual controlling influences.

(k) Summary Exposure to artificial light during the night may disrupt circadian gene function in the

SCN, which in turn may alter circadian-regulated biological pathways, such as cell-cycle regulation and DNA repair. The impact of artificial light on the circadian pacemaker might be modified by genetic variants of the core circadian genes, although such gene–environment interactions have yet to be explored. Given the roles of circadian genes in tumorigenesis, the light-mediated dysfunction of circadian genes may provide a possible


mechanism for the putative carcinogenic effect of light, which may or may not involve melatonin.

4.2.3 Melatonin as part of the neuroendocrine immune axis

Pineal melatonin plays an important part in the neuro-immune-endocrine web regulating mammalian immune defenses. Immune functions in different species of mammals, including man, show circadian and seasonal variations with enhancement during short days, which correlates with the prolonged duration of the daily secretion of melatonin (Nelson & Drazen 2000).

However, in addition to the circadian and seasonally periodic pineal melatonin, the recent observations on the synthesis of melatonin by immune-competent cells in different parts of the immune system suggest a role for locally produced melatonin in the regulation of the immune response. The presence of melatonin and the mechanisms for melatonin-synthesis in many peripheral tissues raises questions about the function of the peripherally produced and/or stored melatonin. There are questions about potential release of this peripheral melatonin into circulation and about its potential participation in circadian system regulation. The relationship of the peripherally formed melatonin to the circadian timekeeping system and its disruptions has not been widely explored, but will be of interest since the same immune-competent cells carry the circadian oscillator genes, and are subject to multifrequency time regulation.

(a) Observations in animals

(i) Pinealectomy – surgical and functional Surgical and functional pinealectomy by continuous bright light exposure led to

abnormal development of the immune organs of mammals and birds (Vaughan et al., 1987; Janković et al., 1994). Impairment of different aspects of the immune response after pinealectomy was reported in rats (Liebmann et al., 1996; Molinero et al., 2000; Beskonakli et al., 2001), in mice (Maestroni et al., 1986; del Gobbo et al., 1989; Vermeulen et al., 1993; Mocchegiani et al., 1996;), other rodents (Haldar et al., 2001), and birds (Rosołowska-Huszcz et al., 1991; Moore et al., 2002; Moore & Siopes, 2003). These defects in immune function could be reversed by the administration of melatonin.

In studies on male Wister rats, constant light which suppresses pineal function and melatonin production induced a 30% depression of the phagocytic ability of blood neutrophils throughout the whole 24-hour cycle without altering its circadian oscillations. It was deduced that the daily dark span serves as synchronizer, and the rhythmic melatonin secretion is involved in the maintenance of the level of phagocytosis and the timing of its circadian rhythm, but does not cause the circadian oscillation as such (Hriscu et al., 2002).

Pharmacological inhibition of melatonin synthesis in mice by the β-receptor antagonist propanolol was shown to be associated with suppressed humoral and cellular

SHIFTWORK 689

immunological responses (Liebmann et al., 1996). Given before onset of the daily dark span, propanolol markedly decreased primary and secondary antibody formation in Balb/c mice injected with sheep red blood cells (Maestroni et al., 1986; Maestroni & Conti, 1996).

(ii) Relation to length of daily photoperiod The relationship between the photoperiod and various aspects of immune function is

stronger with short day lengths (light spans) in diurnal as well as in nocturnal species (Nelson, 2004). There is a correlation between the elevated night time (dark span) melatonin concentrations with the number and response of immunocompetent cells in humans, and in several rodent species (Giordano et al., 1993; Haldar et al., 2001; Prendergast et al., 2003).

(b) Melatonin receptors in immune-competent cells

(i) Observations in animals Immune-competent cells, including splenocytes, lymphocytes and monocytes carry

membrane-bound and nuclear receptors. The membrane-bound MT1 high affinity receptor is coupled to G-protein. The lower affinity MT2 receptor binding sites are not bound to G-protein and have a different pharmacological profile. The melatonin actions in the immune system are mainly mediated by the MT1 receptor. The nuclear receptors found belong to the retinoid-related (retinoid Z receptor/retinoid-related orphan receptors) superfamily of nuclear receptors (Dubocovich, 1995; Dubocovich & Markowska, 2005; Nosjean et al., 2000).

(ii) Observations in humans Specific membrane and nuclear receptors were found in peripheral blood

lymphocytes and monocytes (Lopez-Gonzalez et al., 1992; Pozo et al., 2004). The Kd values of these receptors suggest that they can recognize physiological concentrations of melatonin in circulating blood at night and endogenous melatonin generated locally by the immune system (Carrillo-Vico et al., 2005).

4.2.4 The immunomodulatory response to exogenous melatonin

(a) Observations in animals Melatonin administration in animals leads to immuno-enhancement at several levels

of the immune system ,and in several immune-system-related functions. These actions of melatonin are most pronounced when the animal’s immune system is suppressed, e.g. by light exposure or by corticosteroid suppression (Haldar et al., 2004).

Melatonin administration increased the proliferative capacity of mouse splenocytes (Demas & Nelson, 1998) and rat lymphocytes (Martins et al., 1998), and led to an increase in tissue mass of thymus and spleen (Vaughan & Reiter, 1971; Vaughan et al.,


1987; Rai & Haldar, 2003). The enhancement by melatonin of mouse splenocytes in response to the T-cell mitogen concanavalin A was blocked by the administration of luzindole, a high-affinity melatonin receptor antagonist. Luzindole also reduced the ability of splenocytes to proliferate during the daily dark span (night) when endogenous melatonin concentrations are naturally high. This effect was not observed during daytime (light span) when melatonin concentrations are low (Drazen et al., 2001). The authors suggested that melatonin enhancement of splenocyte proliferation was mediated directly by melatonin receptors on splenocytes and also that the circadian rhythm in splenocyte proliferation was mediated by splenic melatonin receptors (Drazen et al., 2001). Melatonin also acts upon the non-specific immune response and leads to an increase in the number of natural killer (NK) cells and monocytes in the bone marrow (Currier et al., 2000), and enhances the antibody-dependent cellular cytotoxicity (Giordano et al., 1993).

Melatonin in vivo modulates several cytokines active in immune responses via the regulation of their gene expression and production. Among those that are in mice in splenic and/or peritoneal macrophages: the production of tumour necrosis factor α (TNFα), interleukin-1 (IL-1), major histocompatibility complex-II (MHC-II), macrophage-colony stimulating factor (M-CSF), transforming growth factor β (TGFβ), interferon gamma (IFNγ) and IL-10 (Pioli et al., 1993; Liu et al., 2001; Raghavendra et al., 2001a,b). In rats, melatonin increases the generation of thymosin-α1 via an increase in pro-thymosin-α gene expression (Molinero et al., 2000). Interaction between the effects of melatonin upon the immune system with the opiatergic system have been suggested, as in some studies, naltrexone, a specific opioid antagonist, prevented the immune-stimulatory effects of melatonin. Similar effects were observed with the administration of β-endorphin and dynorphin (Maestroni et al., 1988a, Maestroni, 2001).

(b) Observations in humans In humans, exogenous melatonin acts upon NK-cell activity (Lissoni et al., 1986) in a

biphasic pattern. In diurnally active subjects, resting during the night, melatonin given in the afternoon (15:00) led to a stimulation of NK-cell activity during the first 4 hours, followed by a phase of apparent inhibition, suggesting an ultradian periodicity.

4.2.5 Retinal/pineal/hypothalamic/pituitary/adrenal interaction

The relation between the hypothalamic-pituitary-adrenal (Hth-Pit-Adr) axis and the retinal-hypothalamic-pineal (Ret-Hth-Pin) axis are characterized by variables that are rhythmic in several frequencies in each one of these organs (for a review, see Haus, 2007). The optimal function of the mammalian organism depends upon certain time relations between rhythmic variables. In animal experiments, the adrenocorticotropic hormone (ACTH) will elicit a strong adrenal response if given at certain stages of the circadian adrenal cycle in responsiveness when activation of the gland is expected. The response will be substantially less when it is given at other circadian stages both in vivo (Haus, 1964) and in vitro (Ungar & Halberg, 1962; Sánchez de la Peña, 1993). In this

SHIFTWORK 691

regard, the extent of the response of the Hth-Pit-Adr axis to stress (e.g. handling of an animal or saline injection) may be out of phase with the activation of the adrenal gland directly by ACTH due to temporal differences in the cycles of responsiveness of the adrenal and of superimposed controls (Haus, 1964). The rhythmic interactions among hormonal stimuli, rhythmic receptor activity, and target organ response imply that a study of an endocrine and neuroendocrine interaction at a certain (single) time represents a snapshot characterizing only that certain time point. Furthermore, the interactions between the variables studied may be quite different quantitatively and even qualitatively a few hours earlier or later. This is the case in environmentally synchronized organisms and much more so during a phase shift of the organism when different variables have a different time course in phase adaptation (Haus & Halberg, 1969; Fève-Montange et al., 1981). The rhythmicity in the variables studied without a chronobiological experimental design has led to many publications that are difficult to interpret or even contradictory. Studies without such an experimental design have to be qualified as valid only for that specific time and constellation of rhythms of the variables studied.

(a) Melatonin receptors in human central nervous system and pituitary The MT1 melatonin receptor is widely distributed in the human hypothalamus and

pituitary. In addition to the SCN, MT1 immunoreactivity was found in numerous sites including the paraventricular nucleus, periventricular nucleus, supraoptic nucleus, and others. The MT1 receptor was colocalized with some vasopressin neurons in the SCN, and with vasopressin and oxytocin neurons in the paraventricular nucleus, and with parvocellular corticotropin-releasing hormone (CRH) neurons in the paraventricular nucleus (Wu et al., 2006).

The colocalization of MT1 and CRH suggest that melatonin might directly modulate the Hth-Pit-Adr axis in the paraventricular nucleus suggesting cross-modulation between the systems at the hypothalamic level, which may have implications for stress reactions and other conditions (Wu et al., 2006).

In the pituitary, strong MT1 expression was observed in the pars tuberalis while only weak staining was found in the anterior and posterior pituitary (Wu et al., 2006).

(b) Animal studies on pineal-adrenal interaction Both the N-acetyltransferase activity in the pineal glandwhich is the rate limiting step

in melatonin synthesis, as well as adrenal activity are under sympathetic control (Buijs et al., 1999). The rat pineal gland expresses glucocorticoid receptors in a density comparable to the liver (Ballard et al., 1974; Ferreira et al., 2005). The receptors are functional and participate in glucocorticoid-induced effects upon the pineal, which are blocked by the high-affinity glucocorticoid receptor antagonist mifepristone (Gagne et al., 1985). Melatonin binding sites have been identified in the adrenal in rats (Persengiev, 1992), suggesting the existence of a direct interaction between pineal melatonin and adrenal cortex steroid production (Persengiev & Kanchev, 1991).


Glucocorticoids in vitro were reported to decrease the norepinephrine-stimulated melatonin secretion in the rat pineal gland (Fève-Montange & Abou-Samra 1983), and to participate in lowering pineal N-acetyltransferase activity and melatonin production during stress (Joshi et al., 1986).

In perifusion studies on adult rat pineal glands, corticosterone and dexamethasone, but not deoxycorticosterone, decreased melatonin production in pharmacological doses. Lower concentrations had no effect, regardless of the circadian stage (Zhao & Touitou, 1993). Torres-Farfan et al. (2003) studied MT1 melatonin receptors in the adrenal glands of the capuchin monkey, and through sampling at a single time point found inhibition of ACTH-stimulated cortisol production by melatonin. This effect was reversed by the MT1/MT2 antagonist luzindole.

In studying the interaction of the pineal gland with pituitary and adrenal glands in vitro, Sánchez de la Peña et al. (1983a,b) found in a chronobiological study design that the effect of aqueous pineal gland extracts and of melatonin upon production of corticosterone in the mouse adrenal gland did depend critically upon the circadian stage in which the adrenal glands were harvested. In LD12:12-synchronized B6D2F1 mice, pineal gland extracts harvested at one circadian stage and ACTH 1–17, (an analogue of the natural ACTH pituitary hormone), added in vitro to adrenal glands harvested at six different circadian stages showed that, depending upon the circadian stage, when the adrenal glands were obtained, the pineal gland extract or melatonin either stimulated or inhibited or had no effect on adrenal corticosterone production stimulated by ACTH 1–17 (Sánchez de la Peña et al., 1983a, b). The effect of the pineal gland extract or melatonin did depend upon the circadian cycle of responsiveness of the target organ both quantitatively and in direction. The circadian-stage-dependent effect of the pineal gland on the adrenalgland may explain many of the controversial results reported in the literature.

The adrenal androgen dehydroepiandrosterone sulfate (DHEA-S) stimulated melatonin production and secretion by 50–80% in perifused isoproterenol-stimulated rat pineal glands which had been removed during the light phase, while in pineals obtained during the dark span, only the highest doses of DHEA-S increased melatonin secretion, and by only 25% (San Martin & Touitou, 2000). No such effect was observed from dehydroepiandrosterone (DHEA) which is also secreted by the adrenal gland.

In in-vivo studies, a direct inter-relation between the pineal gland and the Hth-Pit-Adr axis could be shown in some models but not in others. DHEA-S, given as single injection in pharmacological doses (500 μg), induced a significant increase in nocturnal pineal gland melatonin content and an increase in N-acetyltransferase in Wistar rats, both young and old. DHEA-S or DHEA at lower doses (50 and 250 μg administered acutely) and at doses of 100 μg administered daily over 8 days had no effect (Djeridane & Touitou, 2004). In a strain of mice with the enzymatic mechanisms to produce DHEA, melatonin stimulated DHEA production in ex-vivo adrenal incubates at all stages of the circadian rhythm (Haus et al., 1996). The administration of tryptophan in rats caused a marked rise

SHIFTWORK 693

in plasma melatonin but had no effect upon corticosterone concentrations (Hajak et al., 1997).

The acute and longer-lasting exposure of rats to stress led to a significant rise in adrenal corticosterone secretion but had no effect upon circulating melatonin levels (Hajak et al., 1997). Some in-vivo models of stress which increase corticosterone secretion such as immobilization (Lynch et al., 1977), forced swimming (Wu et al., 1988) and insulin-induced hypoglycaemia (Tannenbaum et al., 1987) did increase daytime levels of melatonin, and attenuate nocturnal light-pulse inhibition in melatonin synthesis (Funk & Amir, 1999).

In a model of chronic inflammation in rats exposed to Bacillus Calmette-Guérin (BCG), Ferreira et al. (2005) reported in Wistar rats kept on an LD12:12 lighting regimen and sampled at the early part of the light span (09:00–11:00) that corticosterone potentiated noradrenaline-induced melatonin and N-acetylserotonin production in pineal organ culture in a bell-shaped curve through the action of the glucocorticoid receptor. Glucocorticoids exerted a positive control on the secretion of melatonin by the pineal gland in animals undergoing a chronic inflammation process (Ferreira et al., 2005).

When mice were exposed to competing synchronizers (e.g. light versus time-limited meal feeding), the circadian rhythm in corticosteroids tended to counteract the internal desynchronization between central and peripheral oscillators, and tended to stabilize the internal circadian time organization (Le Minh et al., 2001).

In the more complex models, often both the Ret–Hth–Pin and the Hth–Pit–Adr axes react concomitantly. Inconsistencies and controversial results reported in the literature often may be due to the variable constellations seen in one- or two-point snapshots of two or more high-amplitude rhythmic systems.

(c) Human studies on pineal-adrenal interaction

Glucocorticoid secretion was not modified by either acute or chronic melatonin administration in close to physiological doses (Wright et al., 1986; Waldhauser et al., 1987). No correlation was found between the nocturnal urinary excretion of melatonin and cortisol, either among healthy subjects or among patients with panic disorder (with increased excretion of cortisol) or in insomnia patients (with a high incidence of low melatonin secretion) (Hajak et al., 1997).

The circadian rhythms of cortisol and melatonin are related in their timing within the framework of the day–night-synchronized human time organization with activity during the day, and in free-running blind subjects (Skene et al., 1999). The plasma melatonin concentration begins to rise (melatonin onset) when the cortisol concentration is at its lowest, and peaks when the cortisol concentration begins to rise, it then begins to drop (melatonin offset) when the cortisol concentration reaches its peak (Arendt, 1988; Rivest et al., 1989). In case of a rapid phase shift in shiftwork or travel over time zones, melatonin rapidly follows the light–dark and sleep–wakefulness pattern while cortisol


phase-shifts only slowly over a large number of transient cycles (Fève-Montange et al., 1981).

In clinically healthy men, in samples collected every 30 minutes over 24 hours, the ultradian rhythms of cortisol and melatonin followed ultradian periods of about 8 hours and 5.5 hours, respectively (Rivest et al., 1989), suggesting an intrinsic difference in the mechanism controlling their secretion. Similarly, the cortisol and melatonin response to 24 hours of complete bed rest under dim light was also different.

Cagnacci et al. (1995b) administered a high pharmacological dose of melatonin (100 mg) or placebo at 08:00 on two consecutive days to a group of young women (22–32 year of age) in early follicular phase and a group of postmenopausal women (54–62 years of age), and observed gender and age differences in melatonin and cortisol blood concentrations over 48 hours. The postmenopausal group had higher cortisol concentrations than the young group during the daytime (especially at lunch time and early in the evening).

In pituitary- and adrenal-dependent Cushing syndrome with hypercortisolemia patients, the circadian rhythm of melatonin was abolished, and the nocturnal melatonin levels and the integrated 24-hour secretion were significantly lower than in controls (Werner et al., 1981; Soszyński et al., 1989).

In human subjects, melatonin concentrations were markedly reduced after the administration of a low dose (1 mg) of dexamethasone at 22:00 (Beck-Friis et al., 1985). Similar results were reported by Demisch et al. (1988). However, a higher dose of dexamethasone (4 mg given during 1 day in a dosage of 1 mg orally at 08:00, 12:00, 18:00 and 00:00) had no significant effect on melatonin concentration (Beck-Friis et al., 1983 ).

The inhibition of cortisol synthesis with the use of metyrapone resulted in an increase in melatonin urinary excretion (Brismar et al., 1985). No significant difference in cortisol was found during a propanolol-induced decrease in melatonin (Beck-Friis et al., 1983, 1984 ).

The Hth–Pit–Adr and the Ret–Hth–Pin axes are two major branches of the human time-keeping system and provide time information to peripheral tissues. The two axes behave differently during phase shift and phase adaptation. A direct interaction is suggested by the presence of functional receptors for melatonin in the central nervous system, the pituitary gland, and the adrenal gland. However, these interactions, if they are truly functional, may vary with the circadian stage and the responsiveness of the target tissues. At the time of writing, a consistent relationship has not yet been established and many of the studies reported suffer from a lack of a chronobiological study design, which may be required to obtain meaningful results.

4.2.6 Melatonin and the neuroendocrine reproductive axis

Melatonin, as the messenger of darkness in diurnally and nocturnally active species, plays a major role in directing the activity of the reproductive system. It provides input on

SHIFTWORK 695

the length of the daily dark span, thus indicates the season to the species’ neuroendocrine system in the parts of the northern and southern hemisphere where seasonal changes in luminosity occur.

The majority of mammals are seasonal breeders. The role of the pineal gland and of melatonin in controlling mammalian reproduction among seasonal breeders is now well established (e.g. Goldman, 2001). There are differences among different species in the mechanisms elicited by melatonin stimulus, e.g., in the hamster, the reproductive system is inhibited by short photoperiods with a prolonged melatonin signal leading to an anestrous effect in females, and testicular regression in males (Hoffman, 1973; Carter et al., 1982). The anti-gonadal effect of the short photoperiod is prevented by pinealectomy (Reiter 1972, 1980) while gonadal inhibition can be achieved even during long days by the daily injection of melatonin (Stetson et al., 1976). In species like the ewe, the mechanism is different with suppression of the estrous cycle during spring and summer with fertility resumed during autumn and winter (Goldman, 1999, 2001). In this species, there is a different response to the prolonged melatonin signal, leading to a release of gonadotropin (Bittman et al., 1983). Interspecies differences have to be kept in mind if the results of animal experiments are to be applied to humans.

The melatonin effects on the reproductive system in humans and animals appear to take place at several locations, and appear to be receptor-mediated. Melatonin has been shown to exert its reproductive effects at the level of the central nervous system where the presence of melatonin receptors and responsive neurons has been widely demonstrated. Melatonin directly inhibits hypothalamic GnRH pulses (Bittman et al., 1983), and suppresses the pituitary response to GnRH (Martin et al., 1980). However, peripheral actions at the level of the gonads have also been reported. The ovary in experimental animals and humans takes up circulating [3H]melatonin more effectively than most other tissues (Wurtman et al., 1964; Cohen et al., 1978).

(a) Central mechanisms in animal studies Melatonin exerts its regulatory effects on the reproductive axis predominantly through

actions upon the central nervous system (e.g. Glass & Lynch 1981; Lawson et al., 1992; Goldman, 2001). Melatonin binding sites have been demonstrated in numerous areas in the hypothalamus, which are involved in reproductive functions (Migaud et al., 2005; Vanĕcek, 1988; Weaver et al., 1989) together with the pars tuberalis of the pituitary gland (Morgan, 2000). In sheep, the pre-mammillary hypothalamus was identified as the site where melatonin regulates seasonal reproduction (Malpaux et al., 2001; Lincoln, 2002). In times of reproductive quiescence induced by short photoperiods or melatonin treatment, the pituitary gland remains responsive to exogenous GnRH (Robinson et al., 1986) favouring a mechanism located in the hypothalamus rather than the pituitary gland. Animal experiments have shown that the action of melatonin may be mediated by way of regulating gonadotropin release through effects upon hypothalamic monoamines and GnRH (Martin & Sattler, 1982; Arendt, 1986; Reiter, 1991), and by action at the level of the pituitary gland through cAMP- and Ca2+-dependent mechanisms leading to inhibition


of the pituitary response to GnRH (Martin & Klein, 1976; Vanecek, 1995; Vanecek & Klein, 1995). Control of seasonal prolactin operates via the MT1 receptors of the pars tuberalis (Lincoln, 2002, 2006a, b; Lincoln et al., 2003).

In the Syrian hamster, neurons expressing melatonin receptors in the dorsomedial nucleus of the hypothalamus are implicated in the regulation of gonadotropin secretion and gonadal activity (Maywood et al., 1996). In LSH/SsLaK female hamsters, melatonin treatment given approximately one hour before light-off in an L14:D8 regimen decreased significantly the weight of the uterine and pituitary glands, FSH, LH, and prolactin (Lawson et al., 1992). In ovariectomized virgin animals of the same strain, melatonin (25 µg/day subcutaneously) given at the same circadian phase reduced the number of cells expressing estrogen receptor immunoreactivity in the medial preoptic area (Lawson et al., 1992).

At the level of the pituitary gland, melatonin acts in sheep and other photoperiodic animals via MT1 receptors in the pars tuberalis to control seasonal prolactin secretion (Morgan, 2000; Lincoln & Clarke, 1994; Hazlerigg et al., 2001). In the pars tuberalis, it appears that circadian clock genes provide a molecular mechanism by which melatonin duration is decoded (Lincoln et al., 2002; Lincoln, 2006a). The ovine pars tuberalis expresses the core clock genes with a 24-hour rhythm in mRNA levels distinct for each gene and different in timing and amplitude from the clock-gene profiles of the SCN (Lincoln et al., 2002; Hazlerigg et al., 2004). In the pars tuberalis, but not in the SCN, the timing of the clock gene rhythms is markedly modulated by photoperiod and manipulation of melatonin (Hazlerigg et al., 2004; Johnston et al., 2006). Cry1 is controlled in sheep and also in rodents via the MT1 receptor (Hazlerigg et al., 2004; von Gall et al., 2005; Johnston et al., 2005, 2006) with a low amplitude circadian rhythm remaining after melatonin suppression by exposing animals to constant light (Johnston et al., 2006). Under constant light conditions, melatonin was effective at all times in activating Cry1 expression, but suppressed RNA levels for the other clock genes measured (Bmal1, Per1, Per2, Rev-erbα) only at the times when endogenous gene expression was increased (Johnston et al., 2006). A phase-dependence of melatonin action upon the stage of the endogenous rhythms at the level of the target organ may explain many controversial resultsMelatonin onset at dusk activates Cry1 gene expression (the dusk oscillator) and melatonin offset at dawn activates Per1 gene expression (the dawn oscillator), and the interval between these events corresponds to the night length, and thus varies with the seasons. The Per/Cry interval dictates the level of Per/Cry protein complexes in the pars tuberalis cell nucleus achieved during each circadian cycle, and governs the functional output of the pars tuberalis (Lincoln, 2006a, b).

(b) Interactions of estrogen with melatonin at the level of the target organs in animal studies

Estrogens stimulate the growth of ER+ breast cancer cells by stimulating the transcription of cell-cycle progression genes, and downregulating the expression of genes that block cell-cycle progression (Métivier et al., 2003; Stossi et al., 2006). Chromatin

SHIFTWORK 697

remodelling mediated by the estrogen receptor α (Erα) has been suggested as constituting an essential part of mammary tumorigenesis (Sahar & Sassone-Corsi, 2007). Cyclin D1 stimulates mammary growth and in its overexpression leads to mammary tumorigenesis associated with Erα, and enhances its activity by antagonizing the repressor BRCA1 (Wang et al., 2005). Since cyclin D1 is under clock control (Fu et al., 2002), a direct relation between cell proliferation and circadian regulation or dysregulation may play a role in mammary gland cell proliferation and carcinogenesis. Also CLOCK and other linked circadian regulators appear to play a role in cell-cycle regulation, and DNA repair. Actions of CLOCK in its enzymatic functions as enzyme histone acetyletransferase may be involved in chromatin remodelling in response to estrogens in a circadian manner (Sahar & Sassone-Corsi, 2007), and in the case of disrupted circadian rhythms, may lead to alterations in cell proliferation and cancer.

In addition to its central regulatory functions, melatonin has been shown to interfere with the proliferation of human breast cancer cells in vitro (Blask & Hill, 1986; Hill & Blask, 1988). The local inhibitory and anti-estrogenic effects of melatonin have been studied largely in human breast cancer cells of the ER+ and estrogen responsive MCF-7 cell line. Melatonin was shown to downregulate the ER expression in MCF-7 cells, and its anti-proliferative effects appeared to be mediated through the estrogen response pathway (Hill et al., 1992). The melatonin anti-proliferation effect upon breast cancer cells is limited to ERα+ MCF-7 cells and is not found in ERα− (MDA-MB-231) breast cancer cells (Hill et al., 1992). There is substantial literature supportive of an anti-proliferative action of melatonin in physiological concentrations corresponding to peak night time and daytime serum values found in humans, which directly inhibit the MCF-7 cell line in vitro (Blask & Hill, 1986; Hill et al., 1992; Hill & Blask, 1988; Cos & Sánchez-Barceló, 1995). In a recent study, it was shown in a human breast cancer xenograft rodent model that melatonin-rich human blood obtained during night time reduced tumour proliferation while melatonin-depleted blood obtained during daytime or following exposure to bright polychromatic light at night enhanced human breast cancer xenograft proliferative activity (Blask et al., 2005b).

With regard to mechanisms, melatonin suppresses both ERα protein and ERα RNA in a time- and dose-dependent manner (Molis et al., 1994) but does not compete with E2 for binding to the ERα (Molis et al., 1994). In MCF-7 cells, melatonin pretreatment significantly reduced E2-induced ERα transactivation and ERα estrogen-responsive-element-binding activity (Kiefer et al., 2002). Melatonin also inhibited the E2-induced elevation of cAMP levels; melatonin, in this model, acting as biological modifier to affect ERα transcriptional activity (Kiefer et al., 2002).

(c) Prenatal exposure to melatonin in animals Throughout fetal development, expression of the melatonin receptor exhibits

considerable plasticity. During early stages of development, melatonin receptors are transiently expressed in multiple neural and endocrine tissues (Davis, 1997). Expression of MT1 receptors is subject to developmental and circadian control, which may modulate


the physiological actions of melatonin. In studies with cloned regions of the ovine MT1 promoter and studies of the rat promoter, Johnston et al. (2007) suggested a model in which the melatonin expression in the mammalian pituitary gland during development is determined by the changing balance between stimulating and inhibiting transcription factors. In these studies, the authors also suggested that the circadian variation in MT1 gene expression does not depend upon the direct action of circadian clock genes (Johnston et al., 2003a,b, 2006).

In rats, the maternal pineal gland and melatonin (which passes the placental barrier) are necessary for normal sexual maturation. Prenatal melatonin treatment was shown to produce delayed sexual maturation (Díaz López et al., 2005), and hyperprolactinemia in 30-days-old offspring. Melatonin treatment during pregnancy was shown to influence the ontogeny of the hypothalamus–pituitary–gonadal (Hth–Pit–Gnd) axis that begins during intrauterine life, and leads to alterations in gonadotropin and prolactin secretion of both female and male rats during sexual development (Díaz López et al., 2005). The feedback of E2 on LH secretion by the pituitary gland was altered in the female offspring exposed to melatonin in utero, resulting in precocious initiation of puberty. In the male offspring, both the LH and FSH feedback mechanism were delayed. Modification of the fetal endocrine environment caused by prenatal melatonin administration induced changes in the sensitivity of gonadotropin regulation and the prolactin feedback response to exogenous androgens indicative of a delayed sexual development of the male offspring (Díaz López et al., 2005). Increased exposure to melatonin during intrauterine life resulted in an inhibitory effect on postnatal androgen biosynthesis (Díaz Rodríguez et al., 1999). Both maternal pinealectomy and melatonin treatment led to alterations of oocyte development in the female offspring (Fernández et al., 1995).

In the fetus of mother capuchin monkeys (90% of gestation), the MT1 receptor and the clock genes Bmal1, Per2, Cry2 and Clock showed circadian changes in the SCN and adrenal gland, and a rhythm of DHEA-S concentration was found in plasma (Torres-Farfan et al., 2006). Maternal melatonin suppression by a constant light exposure changed the expression of BMAL1, Per2 and MT1 in the fetal SCN. These effects were reversed by maternal melatonin replacement. In contrast to the SCN, maternal melatonin suppression nor its replacement had an effect on the clock genes or MT1 expression in the fetal adrenal gland or the circadian rhythm of fetal plasma DHEA-S. The authors suggested that maternal melatonin is a zeitgeber (synchronizer) for the fetal SCN but probably not for the adrenal gland (Torres-Farfan et al., 2006).

(d) Melatonin effects at the level of the ovary in human studies The uptake of melatonin in animal and human ovarian tissue has been reported

(Wurtman et al., 1964; Cohen et al., 1978). High levels of melatonin which undergo circadian and seasonal variations are foundin human pre-ovulatory follicular fluid (Yie et al., 1995a; Rönnberg et al., 1990). In human granulosa-luteal cells, melatonin binding sites have been detected (Yie et al., 1995b), and a stimulation of progesterone production by melatonin has been shown (Brzezinski et al., 1992; Webley & Luck, 1986). Several

SHIFTWORK 699

forms of melatonin receptor genes are expressed in human granulosa-luteal cells (Niles et al., 1999). Woo et al. (2001) identified the melatonin receptor subtypes MT1-R and MT2-R. Cloning and sequence analysis revealed that these receptors were identical to their brain counterparts. Treatment of these cells with melatonin significantly increased the LH receptor mRNA levels without any effect on the expression of the FSH receptor gene. After melatonin treatment, both GnRH and GnRH receptor mRNA were significantly decreased to 61% and 45% of control levels, respectively. In the same study, melatonin itself had no effect upon basal progesterone production, but enhanced human chorionic gonadotropin (hCG) stimulated progesterone production. There appeared to be a complex receptor-mediated direct melatonin action upon ovarian steroidogenesis involving the LH and GnRH receptor gene expression in the steroid-producing human granulosa-luteal cells. These peripheral actions of melatonin complement its central actions and can conceivably lead to an alteration of the gonadal steroid balance (Woo et al., 2001).

(e) Melatonin during puberty in humans Serum night time melatonin concentrations are high in children, and drop by 75%

from childhood (1–5 years) to young adulthood (Waldhauser et al., 1984; Waldhauser & Dietzel, 1985). The morning values are uniformly low without change over different ages. The night time melatonin concentration were negatively correlated with the Tanner Stages of sexual development. In contrast, the aMT6s excretion in children and adults was similar in per day amount (Tetsuo et al., 1982; Bojkowski et al., 1987b). It appears that the amount of melatonin secreted by the pineal gland from childhood to young adulthood remains about the same, but as it is distributed over a larger body volume, the serum concentration is lower.

(f) Melatonin during the menstrual cycle in humans Conflicting results have been reported in studies of the circadian melatonin rhythm

during the menstrual cycle. Some studies reported increased melatonin levels during the luteal phase (Wetterberg et al., 1976; Arendt 1978, 1988; Webley & Leidenberger, 1986; Brun et al., 1987) or no difference between the phases (Brzezinski et al., 1988; Wright et al., 2001).

Brzezinski et al. (1988) found no significant change of plasma melatonin during the normal menstrual cycle in 14 clinically healthy normally cycling women (± 36 years of age (range 19–34)) studied at 2-hour intervals over a 24-hour span during early follicular, periovulatory, and luteal phase of the menstrual cycle. Circadian phase, amplitude, and total amount of melatonin secreted were consistent among the three profiles.

Studying the relations of melatonin to FSH and LH in 79 healthy women of different ages, Fernández et al. (1990) found a significant correlation of melatonin with FSH and E2 in menstruating women during the follicular phase, while during the luteal phase, a negative correlation was found between melatonin, progesterone, and E2. During the perimenopausal period, there was no significant correlation between the serum hormone


concentrations. In menopause, as during the follicular phase, melatonin and FSH were negatively correlated.

Exposure to bright light at night appears to have some effects upon the regulation of the menstrual cycle (Dewan, 1967; Lin et al., 1990). A response of menstrual cycle length to nocturnal light exposure (100 W bulb with 235 lux) has been reported in women with long and irregular menstrual cycles (Lin et al., 1990). Nocturnal light may also have effects upon the menstrual cycle phase (Putilov et al., 2002).

There is no direct evidence that either endogenous, nocturnal circulating levels of melatonin or the administration of exogenous doses of melatonin mimicking circulating physiological concentrations exert any influence on pituitary gonadotrophins, prolactin, or gonadal steroids in humans. However, exposure of normal menstrual cycling women to continuous light (500–800 lux measured at eye level) during the night suppressed nocturnal circulating melatonin and prolactin concentrations while elevating FSH concentrations; no clear-cut effects were observed on LH levels when compared to control subjects maintained in the dark (Miyauchi et al., 1991). This same group later reported that the incidence of menstrual irregularities in a cohort of 766 women who worked in various occupations was highest in nurses (24.9%), factory workers (36.8%), and barmaids (40.0%) when compared to teachers (13.1%) and office personnel (14.9%); the incidence of menstrual irregularities were significantly higher in those working during the night versus the day. In a small subset of nurses, those working during the night (n = 5) had significantly lower blood concentrations of melatonin and prolactin (sampling at 22:00 and 02:00) versus nurses resting in their quarters (n = 6); however, no differences were observed in plasma LH or FSH levels (Miyauchi et al., 1992). In a study of 53 healthy women exposed to light during the night, circulating melatonin levels during the night were suppressed while there was no point for point changes in matching mesures of circulating E2 levels regardless of whether women were in the follicular or luteal phases of the menstrual cycle. Furthermore, in women who chronically secrete low or high levels of melatonin during the night (area under the curve range) had similar E2 blood levels (Graham et al., 2001). This was also true in nude female rats exposed to increasing intensities of white, fluorescent light during the dark phase (0, 0.02, 0.05, 0.06, 0.08 and 345 μW/cm2) of an LD12:12 regimen – a dose-dependent suppression of the nocturnal amplitude of blood melatonin levels was observed while circulating levels of E2 were unchanged (Blask et al., 2005b)

Disruption of circadian rhythms is associated with disturbances in menstrual function. Female shiftworkers compared to non-shiftworkers are more likely to report menstrual irregularity, and longer menstrual cycles (Baker & Driver, 2007).

Menstrual cycle irregularities have been reported in female airplane crew members, which may be the result of frequently repeated phase shift, light exposure at times unusual for their circadian cycle, or other causes specific to air travel (Iglesias et al., 1980). The frequent phase shift in airline personnel has also been reported to lead to cognitive deficits (Cho et al., 2000) and even associated with organic changes in the temporal lobe area

SHIFTWORK 701

(Cho, 2001). No observations on reproductive axis dysregulation were mentioned in these reports

(g) Seasonal variations of pineal-ovarian relations in humans In human subjects, the availability and exposure to artificial illumination appears to

decrease the seasonal differences in environmental daily light–dark span, and the associated changes in pineal–gonadal relation. However, in Northern countries with strong seasonal variation in luminosity, melatonin also seems to contribute to the seasonal control of reproductive function in humans. During the dark months of the year, the activity of the pituitary-ovarian axis (Ronkainen et al., 1985) on the conception rate (Timonen et al., 1964; Sandahl, 1978) is decreased.

Studying serum melatonin as the likely messenger of the length of the daily dark span, Kivelä et al. (1988) found that the serum melatonin concentrations on menstrual cycle Days 2 and 10 of 12 clinical healthy, diurnally active women were 27% and 49%, respectively, higher in the winter than in the summer. Night time serum LH levels at mid-cycle were 76% higher in the summer than in the winter. The high levels of melatonin in the winter may have had an inhibiting effect upon serum LH levels (Kivelä et al., 1988).

Kauppila et al. (1987) found that the area-under-the-curve of 24-hour melatonin profiles obtained by 2-hourly serum sampling during the dark (winter) season in 11 normally cycling women were significantly larger than during the light season. The duration of the nocturnal melatonin pulse during the dark season was lengthened whereas the mean serum E2 concentration was significantly decreased at the time of ovulation and during the luteal phase of the cycle in spite of increased gonadotropin concentration, indicating lowered ovarian responsiveness. The concentration of free testosterone was also lower during the dark season.

(h) Aging of the pineal-gonadal axis in the rat In middle-aged female rats (11-months-old) with irregular estrous cycles and lowered

gonadotropin surge during proestrus (perimenopausal in human equivalent), melatonin enhanced the amount of LH, FSH, and prolactin released during the surge at the proestrus day and restored the afternoon preovulatory surge in LH, FSH, and prolactin to values equivalent to those found in young rats. E2 concentrations were markedly increased in the treated animals on the day of proestrus which preceded the FSH, LH, and prolactin surges in the afternoon (Díaz et al., 1999). Melatonin administration in middle-aged female rats regulated the activity of the hypothalamo-pituitary unit, and particularly improved gonadotropin secretion in response to the luteinizing-hormone-releasing hormone (Díaz López et al., 2005; Díaz et al., 1999, 2000).

In acyclic 23–25-months-old rats, melatonin reduced the elevated LH and FSH concentrations and increased the prolactin concentration (Diaz et al., 2000). The responsiveness of the pituitary to the luteinizing-hormone-releasing hormone in vivo was increased by melatonin treatment, which in aging animals restored the pituitary


responsiveness to levels similar to that seen in young rats (Díaz et al., 1999; Fernández Alvarez et al., 1999). This is in contrast with in-vitro findings in the neonatal pituitary gland (Martin & Klein, 1976). Therefore, the effect of melatonin changes with the age of the animals.

(i) Menopause in humans Under controlled ‘constant routine’ conditions, there was no significant difference in

the amplitude of the salivary melatonin circadian rhythm between healthy middle-aged premenopausal (age 42 ± 4 years) and postmenopausal (55 ± 2 years) women (Walters et al., 2005). In this respect, this study agrees with the findings of Zeitzer et al. (1999), which showed no age-related difference in melatonin amplitude when subjects were studied under constant routine conditions (although the age difference of the groups in this study was small).

There was a significant advance of the timing of the melatonin acrophase in the postmenopausal compared to the premenopausal women (1.1 ± 0.5 hours versus 2.3 ± 0.3 hour clock time in decimals, respectively) (Walters et al., 2005). This result is in agreement with the studies of Cagnacci et al. (1995b), Duffy et al. (2002) and Yoon et al. (2003) while other investigators reported a phase delay (Sharma et al., 1989), or no change in timing (Youngstedt et al., 2001).

The correlation between melatonin and LH and melatonin and FSH was negative in perimenopausal and menopausal women before treatment with melatonin. After 6 months of treatment with 3 mg of melatonin at bedtime, a significant decrease in plasma LH was found only in the younger women (43–49 years of age) and not in the older women (50–62 years of age). There was a significant decrease of FSH especially in women with low basal overnight melatonin levels. In the same study, the women treated with melatonin had a significant increase in concentrations of total thyroid hormones, triiodothyronine (T3), and thyroxine (T4) in comparison to the women treated with placebo, without concomitant changes in the thyroid-stimulating hormone (TSH) on single-time-point sampling (Bellipanni et al., 2001).

(j) Melatonin in disorders of the human hypothalamic–pituitary–gonadal axis

(i) Women athletes with functional hypothalamic amenorrhoea Women athletes have abnormalities of the hypothalamic–pituitary–ovarian (Hth–Pit–

Ova) (Veldhuis et al., 1985; Loucks et al., 1989) and hypothalamic–pituitary–adrenal (Hth–Pit–Adr) axes (Loucks et al., 1989; Ding et al., 1988). The changes in the Hth–Pit–Ova axis resemble those of women with functional hypothalamic amenorrhoea who do not exercise (Berga et al., 1989; Suh et al., 1988) in whom the magnitude and duration of nocturnal melatonin secretion is increased. Elevated daytime plasma concentrations of melatonin were observed in cycling and in amenorrhoeic women athletes compared to sedentary women. In contrast, nocturnal melatonin concentrations in sedentary and

SHIFTWORK 703

cycling athletic women were indistinguishable while the amenorrhoeic athletic women had a marked increase in nocturnal peak amplitude and delay in melatonin offset leading to a 2-fold amplification of the nocturnal melatonin secretion (Laughlin et al., 1991). Neither opioidergic (naloxone) nor dopaminergic (metoclopramide) blockade changed melatonin secretion in any of the three groups. The mechanisms of the amenorrhoea in the athletes appeared to be similar to that of sedentary functionally amenorrhoeic women. These mechanisms seemed related to a common hypothalamic dysregulation rather than to athleticism which was accompanied by daytime elevated values of melatonin, and not by the characteristic elevation seen in the amenorrhoeic subjects during night time (Laughlin et al., 1991).

(ii) Functional hypothalamic amenorrhoea Plasma melatonin concentrations in seven women with functional hypothalamic

amenorrhoea (aged 22–35 years, mean 28.4), measured in 2-hourly sampling over a 24-hour span (daytime and night time), were significantly higher than in concomitantly studied healthy controls observed during three stages of their menstrual cycle (Brzezinski et al. 1988). Similar results were reported by Berga et al. (1988) in seven women with the same condition sampled at 30-minute intervals over a 24-hour span. While the daytime melatonin levels were undetectable in both groups, the integrated night-time levels were three times greater in the amenorrhoeic women than in the matched controls. The rise was due both to an increased peak amplitude and an extended duration of melatonin secretion towards morning, in spite of a comparable light–dark regimen. In a further study by Okatani & Sagara (1994), 20 women with functional hypothalamic amenorrhoea had significantly higher nocturnal melatonin concentrations than 11 matched controls. Negative correlations between the cumulative melatonin concentration (between 20:00 and 08:00) and serum E2, and between the peak serum melatonin values and serum E2 were observed. Intravenous administration of conjugated estrogen (Premarin 20 mg) significantly suppressed nocturnal melatonin secretion.

Five women with endometriosis and a low estrogen state induced by a GnRH agonist treatment had an increase in nocturnal melatonin secretion which was similar to that of the women with hypothalamic amenorrhea. This observation suggests that melatonin does not alter gonadotropin responses in humans to GnRH (Weinberg et al., 1980; Fideleff et al., 1976).

(iii) Hth–Pit–Gnd disorders in the male Hypogonadotropic hypogonadism and delayed puberty are based on GnRH

deficiency. Both of these conditions in young males resulted in a marked increase in nocturnal melatonin concentrations and integrated nocturnal melatonin secretion values (area-under-the-curve) when compared to normal pubertal male controls (Luboshitzky et al., 1995). There was no correlation between melatonin and LH or between melatonin and prolactin concentrations, suggesting that circulating sex steroids rather than LH modulate melatonin secretion in a reverse fashion (Luboshitzky et al., 1995). This nocturnal


increase in melatonin secretion was corrected using a testosterone treatment resulting in the melatonin concentrations returning to normal levels (Luboshitzky et al., 1996a). In a single case report of female delayed puberty treated with estradiol, a marked decrease of melatonin production was noted in the course of treatment (Arendt et al., 1989).

In contrast, untreated males with hypergonadotropic hypogonadism, due to untreated Klinefelter syndrome and sampled overnight, all had elevated FSH, LH, and E2 concentrations. Klinefelter syndrome patients with low testosterone concentrations had significantly lower melatonin nocturnal concentrations and area-under-the-curve profiles when compared to Klinefelter syndrome patients with normal testosterone levels, and controls. No correlations between the melatonin concentration and LH, FSH, or E2 levels were observed (Luboshitzky et al., 1996b).

In male and female patients with central precocious puberty with elevated sex steroid levels, serum melatonin levels were lower than normal subjects (Waldhauser et al., 1991, 1993) indicating that in general, in conditions in which sex hormones are lower with decreased or normal gonadotropin concentrations, melatonin was found to be elevated, and vice-versa. (Berga et al., 1988; Brzezinski et al., 1988; Laughlin et al., 1991; Tortosa et al., 1989; Okatani & Sagara 1994; Puig-Domingo et al., 1992).

4.2.7 Time organization in the normal and abnormal human breast

(a) Periodicity in the normal human breast Periodicities in the normal human breast were studied non-invasively by measuring

the skin temperature above the mammary gland by ambulatory thermography monitoring. Thermography methods study both the metabolic activity of the gland and the blood flow in the overlying tissue. In some studies, the thermography readings over the breast were related to simultaneously monitored skin temperatures obtained at other sites or to oral temperature allowing to partly separate these two components, and obtain a corrected “breast-specific temperature” (Simpson et al., 1989). Circadian (about 24-hour), circaceptan (about 7-day), and circamenstrual (about 28–32 day) rhythms were identified (Gautherie & Gros, 1977). The breast temperature in clinically healthy diurnally active women exhibited a circadian rhythm similar to that of the oral or the core temperature with a peak (acrophase) in the evening (Mansfield et al., 1973; Gautherie & Gros, 1977). The circamenstrual rhythm in breast temperature included changes in the circadian rhythm parameters in mean amplitude, and a characteristic periovulatory rise with a peak occurring approximately 24 hours after ovulation (Phillips et al., 1981; Wilson et al., 1983). In large numbers of breast biopsies taken from women in different phases of the menstrual cycle, Anderson et al. (1982) found the highest incidence of epithelial mitoses to occur before the onset of the menstrual period.

SHIFTWORK 705

(b) Time of initiation of breast cancer The initiation of breast cancer may occur many years before the clinical manifestation

of the tumour. In many instances, a breast cancer may begin to develop in the early premenopausal period (Simpson et al., 1988). Findings concerning the influence of age at first pregnancy (MacMahon et al., 1970) and the relation of the time of radiation exposure to the appearance of the tumour (Howe, 1984) support the concept that the predominant environmental initiation of breast cancer occurs in the premenopausal period during the reproductive lifespan. This is when the epithelium is proliferating and biologically vulnerable to carcinogenic agents interfering with the circadian periodicity of cell proliferation (Simpson et al., 1988).

An increased cancer incidence in shiftworkers may be related both to initiation of the tumour and to events occurring during the period of promotion of the malignancy, until it becomes clinically manifest.

(c) Circadian time structure and risk to develop breast cancer Gautherie and Gros (1980) reported on a large series of women who received routine

breast examination, including a breast thermogram. Of these, 1245 were followed up for a period of 12 years because of a questionable abnormal thermal pattern. During the follow-up period, 501 of these developed breast cancer.

In a group of 106 women with apparently healthy breast and no family history of related medical conditions, but with an abnormal thermal pattern, 27.2% developed breast cancer. In a group of 31 women with family history for breast cancer and an abnormal thermogram, 11 women (35.8%) developed breast cancer when compared to only 3.9% of 486 women with a normal thermogram. The abnormal thermal pattern preceded the clinical diagnosis of cancer in the majority of cases by 4 to 5 years, but in some instances beyond 5 years, but usually less than 10 years (Gautherie, 1983; Amalric et al., 1981).

Rhythm alterations in the circadian timing of 12 hormonal variables were reported in women with a high risk (epidemiologically determined) of developing breast cancer (Ticher et al., 1996). A total of 24 clinically healthy, diurnally active non-obese American women of three age groups (adolescent 17 ± 2 yr, n = 8; young adult 33 ± 1 yr, n = 10; and postmenopausal 56 ± 7 yr, n = 8) were studied. The women were characterized as high risk (n = 12) or low risk (n = 12) of developing breast cancer according to the epidemiological index criteria. Risk assessment followed a scale based upon the epidemiological data presented by MacMahon et al. (1973), Farewell et al. (1977), and Choi et al. (1978), which are similar to another review of this topic (Gail et al., 1989). No subjects with a family history of BRCA1 or BRCA2 mutations were included. A family history of first degree relatives with (sporadic) breast cancer was the primary distinction, as it applied to all age groups. No medications known to affect prolactin secretion, including oral contraceptives, were allowed 6 months before the start of the study until its completion.


Most subjects were sampled throughout a series of four 24-hour spans during a single year, once each season, and at a different menstrual stage. There were no seasonal differences in the incidence of the different stages of the menstrual cycle. The total number of time series analysed was 44 for the cohort of high-risk subjects (31 in regularly adult menstruating women and 14 in postmenopausal women) and 41 for the cohort of low-risk subjects (26 in regularly non-menopausal menstruating women and 15 in postmenopausal women). The number of time series analysed per season was as follows: n = 21 spring, n = 20 summer, n = 20 autumn, n = 24 winter. Blood for cortisol and prolactin was collected every 100 minutes over each 24-hour span, and blood for the other variables (aldosterone, cortisol, DHEA-S, E2, insulin, LH, 17-hydroxyprogesterone, TSH, thyroxine, and triiodothyronine) every 100 minutes. The data for each subject were analysed for each variable by the single cosinor test (Nelson et al., 1979) yielding a calculated peak time (acrophase), an amplitude, and a circadian-rhythm-corrected mean value (MESOR). The differences and similarities between the high- and low-risk groups and the age groups in the dispersion of the set of acrophases and the ratio of amplitude over MESOR were analysed by a multiple Pearson correlation test and the resulting correlation matrix was used for cluster analysis. Two main profiles of acrophase dispersion were detected according to the level of breast cancer risk. The circadian time organization was similar in women with a high risk to develop breast cancer, irrespective of age, and different from the pattern in women with a low risk. In contrast, the amplitude/MESOR ratio was characteristic for the age group, and unrelated to breast cancer risk (Ticher et al., 1996).

In the same study, Lewy et al. (2007) compared the distribution of circadian and ultradian (in the range of 4–18 hours) rhythms in low-risk and high-risk patients sampled during the four seasons for prolactin and cortisol. The high-risk and low-risk patients expressed different rhythmic output patterns in both variables, also suggesting that the genetic background as defined by the risk state to develop breast cancer was related to differences in the circadian time structure including the ability to change the subjects’ predominant rhythm periods as a function of season. The low-risk patients exhibited a statistically significant change in the rhythm periods of both variables with a shift from the circadian to an ultradian rhythmicity as a function of the season while the high-risk patients did not.

Rhythm alteration in the menstrual temperature rhythm of patients at high risk to develop breast cancer was described by Simpson et al. (1989). The high-risk state in this study was defined by previous excision of an ipsilateral or contralateral breast tumour. While the basic breast-specific temperature in the women at usual risk of breast cancer showed the usual variation characteristic for the menstrual period with a sustained rise after ovulation and high values during the luteal phase, the high-risk patients had three temperature peaks separated by 7 and 6 days, respectively, the largest (first peak), preceding the salivary progesterone peak by about 6 days, the second and the third peaks appearing 2 days and 8 days after the salivery progesterone peak, respectively (progesterone peak appearing 8 days after the ovulation).

SHIFTWORK 707

These data indicate significant changes in the circadian and menstrual (possibly adaptive) characteristics in the human time structure related to the risk state to develop breast cancer.

4.2.8 Sleep deprivation – impact upon the neuroendocrine and immune system

The most prevalent health problem for the night worker and shiftworker is the quantity and quality of sleep. The night worker, but also the early-morning shiftworker, sleeps about >2 hours less than the average day worker, and often with decreased sleep efficiency and sleep quality. Sleep deprivation impacts heavily upon the entire neuroendocrine-immune system complex regulating several biological functions, including cell proliferation, immune defence and adaptation, and defence to everyday stresses.

(a) Sleep deprivation and the immune system The immune system in all its components is closely integrated in two-way

communication and feedback loops with the nervous and the endocrine system, forming a web of biological regulation, which functions rhythmically in multiple frequencies. In the circadian frequency range, the immune system is subject to the central hypothalamic oscillator in the SCN with peripheral oscillators in immunocompetent cells and organs. It is kept in pace by neural as well as neuroendocrine and endocrine messengers and synchronizers. Some of the multifrequency periodic neuroendocrine variables (e.g. prolactin, melatonin) enhance immune reactions, while other variables (e.g. cortisol) keep them in check and control their intensity, or if these variables are overexpressed, they might suppress the immune response.

Immune reactions taking place in mammalian organisms (including humans) exert feedback effects upon the regulatory centres. The immunocompetent cells involved in an immune reaction produce humoral messengers that act on the neuroendocrine system signalling the occurrence of damage and/or of an ongoing immunodefence reaction. The feedback mechanism from the peripheral cells to the centre elicits a neuroendocrine response, which in turn regulates the peripheral cellular response to the stimulus encountered. While cortisol acts as an immunosuppressor, melatonin enhances IFNγ and IL-1 production (Maestroni et al., 1986; Caroleo et al., 1992; Colombo et al., 1992; Morrey et al., 1994), and antagonizes the immunosuppressive effect of cortisol (Maestroni et al., 1988a, b). There is a bidirectional link between sleep and the immune system, in which cytokines such as IL-1, IL-2, interferon, and TNF induce sleep (Krueger & Obál, 1993).

In addition to cytokines, human peripheral leukocytes, e.g. infected by a virus or exposed to endotoxin will synthesize immunoreactive ACTH, and endorphins. The immunoreactive ACTH produced by the immunocompetent cells appears to be identical to pituitary ACTH, and acts upon the same receptors in the target tissues and shows a steroidogenic response in mice. The production of ACTH, both by pituitary cells and by


leukocytes in response to synthetic corticotropin-releasing factor (CRF), is suppressed by dexamethasone in vitro and in vivo suggesting that the production of ACTH and endorphins by leukocytes is controlled by the CRF (Smith et al., 1986). CRF in itself has anti-inflammatory effects that are independent of the pituitary and adrenal glands (Kiang et al., 1987; Gao et al., 1991; Wei & Gao 1991; Serda & Wei 1992). CRF is produced in the hypothalamus in a cyclic fashion, and also in response to a wide variety of environmental stimuli as a stress response and in response to pro-inflammatory cytokines such as IL-1 (Anderson et al., 1993; Rivier, 1993). In animal experiments, elimination of cyclicity by exogenous CRF leads to altered response patterns after challenge (e.g. by bacterial endotoxin) (Linthorst et al., 1997).

The alternation of sleep and wakefulness is a fundamental part of the organization of circadian time and coordinates numerous neuroendocrine variables. Several immunologically active hormones and peptides influence the sleep–wake cycle of the brain and are involved in a bidirectional or multidirectional communication between neuroendocrine, immune, and central nervous system functions. Most of those possess circadian rhythmicity, and some are influenced by environmental factors acting as synchronizers or as masking agents. IL-1 and other cytokines play a regulatory role on the sleep–wake cycle (Opp et al., 1992), and interact in this process with various immunologically active neuroendocrine substances (Krueger et al., 1990a,b). In the physiological regulation of the sleep–wake rhythm, the natural sleep-promoting or wakefulness-promoting substances require a specific constellation of multiple variables that in part is determined by their circadian rhythms and in part by environmental and behavioural conditions (Moldofsky, 1994). Periods of predisposition to sleepiness are separated by periods of resistance to sleep (Lavie & Weller, 1989).

Sleep deprivation in experimental animals and in human subjects leads to an impairment in immune function which, if prolonged, will lead to the death of the animals and of the human subjects (fatal familial insomnia syndrome) (Everson, 1993; Portaluppi et al., 1994). Sleep deprivation in rodents, even for a brief 7-hour period, leads to a downregulation in immune defence against viral infection, and after challenge, to significantly decreased antibody titres (Brown et al., 1989a, b). With more prolonged sleep deprivation, Everson (1993) reported in rats a breakdown in immune defences with a systemic infection by pathogenic organisms, leading to the death of the animals.

Phase shifts, as they are encountered in shiftworkers or after transmeridian flights or rhythm disturbances under irregular work schedules, lead to internal desynchronization of immune-related circadian rhythms, and to the impairment of immune functions. In studies on shiftworkers, Nakano et al. (1982) reported lower proliferative responses in lymphocytes when compared to regular daytime workers. As shiftwork is usually accompanied by a certain degree of sleep deprivation, it is unclear whether the impairment of immune function in shiftworkers is a consequence of circadian desynchronization, of sleep deprivation, or of both. In night shiftworkers, a short daytime sleep isa consequence of circadian desynchronization as a result of the misalignment of

SHIFTWORK 709

the circadian rhythm in sleep propensity of the worker with the time for sleep allowed by the work schedule.

(i) Total sleep deprivation Unfortunately, many of the studies in this area have limited their sampling to single or

very few time points per day, which, in view of the high amplitude circadian rhythms of immunocompetent cells and their responsiveness to stimulation, raises questions as to their interpretation since they often do not allow a distinction between an actual change in level of the variable studied or a shift in circadian phase (Haus, 1996; Haus & Smolensky, 1999). This problem is compounded by different sampling times used in different studies. In prolonged studies of sleep deprivation (e.g. 64 hours) with measurements of circulating immunocompetent cells limited to single time points at a given clock hour only, biphasic or other reaction patterns may be an expression of circadian phase alteration level changes or masking. These sampling limitations in much of the published literature are a likely explanation of the many contradictory results obtained by different investigators.

After 48 hours of sleep deprivation, with blood sampling at single time point at 08:00 before and after 24 and 48 hours of sleep deprivation, and 24 hours after recovery sleep, Ozturk et al. (1999) found a decrease in the proportion of NK cells during sleep deprivation with return to normal after recovery sleep. In contrast, Dinges et al. (1994) sampled blood daily at 22:00 in 20 healthy young adults of both genders over a 64-hour span of total sleep deprivation and at 22:00 in the pre-deprivation day and on the first recovery day. They reported during total sleep deprivation an increase in the number of circulating white blood cells, including granulocytes, monocytes, and NK cells as well as an increase in NK-cell activity, and increased response of lymphocytes to phytohaemagglutinin, a T-cell mitogen. On the basis of their observations, Dinges et al. (1994) assumed that an activation of these branches of the immune system occurred with 64 hours of total steep deprivation.

Seventy-seven hours of total sleep deprivation in eight clinically healthy women led to reduced phagocytosis by polymorphonuclear granulocytes, increased interferon production by lymphocytes, and increased the plasma cortisol concentration (Palmblad et al., 1976). Sampling was limited to a single time point at 12:30, and after 28 and 76 hours of total sleep deprivation. The experimental design included a stressful surrounding during sleep deprivation with simulated battlefield environment.

Also using a single time point of measurement in 12 healthy young men after 48 hours of total sleep deprivation and in-vitro phytohaemagglutinin stimulation of their lymphocytes, Palmblad et al. (1979) reported a decreased lymphocyte blastogenesis.

These differences in the outcome of single-time point studies of variables with high amplitude circadian rhythms emphasize the need for studies at more than one circadian stage, or the use of a marker rhythm (Haus et al., 1988) when studying conditions like shiftwork in which circadian phase alterations are to be expected.


(ii) Partial sleep deprivation In contrast to prolonged total sleep deprivation, partial sleep deprivation with

wakefulness either during the first part or the latter part of the night corresponds more closely to the situation encountered by the shiftworker.

Irwin and colleagues (1994) studied NK-cell activity in 23 healthy adult men (22–61 years of age) with sampling at a single clock hour (between 07:00 and 09:00) after one night of late night partial sleep deprivation, with no sleep from 03:00 to 07:00 following three baseline nights of regular sleep, and again after a recovery night of sleep. The late night sleep deprivation was associated with a decreased NK-cell activity in 18 of the 23 subjects with an average reduction in lytic activity of 28%.

In a similar study of early night of partial sleep deprivation with no sleep until 03:00, Irwin et al. (1996) found in 42 clinically healthy men after a single night by single-time point sampling at the same clock hour (between 07:00 and 09:00), a reduction of the natural immune response as expressed by a decrease in NK-cell activity, NK activity per number of NK cells, decrease in lymphokine-activated killer cell number and activity, and lymphokine-activated killer activity per number of lymphokine-activated killer precursors. IL-2 production stimulated by concanavalin-A was also suppressed. After one night of recovery, sleep NK-cell activity had returned to baseline levels while IL-2 production remained suppressed. These data indicate that even a modest disturbance of sleep produces a reduction in natural immune responses and T-cell cytokine production.

(iii) Sleep deprivation and cytokine balance The immune system is organized in a two-branch model. The pro-inflammatory Type

1 T-helper1 (Th1) cytokines (IL-2IFNγ and IL-12) produced by immunocompetent peripheral blood mononuclear cells is counterbalanced by the anti-inflammatory group of Type 2 T-helper2 (Th2) cytokines (IL-4 and IL-10) (Lucey et al., 1996). In diurnally active human subjects, the Type 1 immunodefence pattern predominates during night hours (Petrovsky, 2001; Dimitrov et al., 2004a). The Type 1 cytokines support the cellular aspects of immune response and are moderated in their pro-inflammatory action by the Type 2 cytokines. Maintenance of this balance is essential since excessive production of one or the other type of cytokines leads to immune disturbances with either inflammation and tissue damage or with susceptibility to infection and allergy (Lucey et al., 1996). The balance among the cytokine groups is maintained in the healthy organism by cross-inhibition and by superimposed neuroendocrine control (Romagnani, 1996), which in its time organization is directed by the SCN, and the related circadian clock mechanisms.

Sleep is an integral part of the circadian time structure and plays a vital role in the regulation of the immune system (Bryant et al., 2004). There is a sleep-associated shift towards Type 1 cytokine activity in T-cells (Dimitrov et al., 2004a). Sleep deprivation leads to alterations in the cytokine balance. A shift towards Type 2 activity has been reported for sleep deprivation in otherwise healthy subjects, in insomnia, under stress, and in the aged (Dimitrov et al., 2004b; Sakami et al., 2002–2003; Glaser et al., 2001).

SHIFTWORK 711

Elevation of sympathetic tone during the night also contributes to a reduction of cellular immunity, e.g. in psychological stress situations (Irwin et al., 1990a; Irwin et al., 1991).

Covering an entire 24-hour span with frequent sampling and sleep permitted from 23:00 to 07:00, and a second 24-hour span without sleep in the same subjects at least 4 weeks apart, Lange et al. (2006) studied by flow cytometry IL-12- and IL-10-producing monocytes, representing messengers of the Th1 and Th2 pattern, respectively. During sleep, there was an increase in the number of IL-12- producing monocytes and a concurrent decrease of IL-10-producing monocytes, leading to a circadian rhythm in these cells with a peak at 02:20 and 11:30, respectively. These apparently rhythmic temporal variations were absent during continuous wakefulness. Monocytes are a major contributor to pro-inflammatory cytokine production in the peripheral blood. Nocturnal sleep shifts monocyte cytokine production to Type 1 cytokines, which is regarded as a prerequisite for sleep-associated predominance of Th1-mediated adaptive immune defence (Lange et al., 2006). The human monocytes are regarded as direct precursors of antigen-presenting cells, and can be directly assessed by flow cytometry in blood samples (Geissmann et al., 2003). The study of Lange et al. (2006) shows a dependence of the cytokine rhythm on sleep and its apparent absence during continuous wakefulness. The circadian variation in monocyte-derived IL-12 and IL-10 production, and the respective Type 1/Type 2 cytokine balance, which are induced primarily by sleep, are vulnerable to sleep disturbances and sleep deprivation.

With regard to mechanisms, growth hormone and prolactin shift the Type 1/Type 2 balance towards Type 1, whereas cortisol and norepinephrine shift it towards Type 2 (Dimitrov et al., 2004a, b; Elenkov & Chrousos, 2002). Both prolactin and growth hormone rhythms are altered during sleep deprivation (Lange et al., 2006). There was a positive correlation between the prolactin level and IL-12+ monocyte numbers, and between norepinephrine and IL-10+ monocyte numbers, and a negative correlation between the cortisol level and IL-12+ monocyte numbers (Lange et al., 2006; Petrovsky & Harrison, 1998). In vitro, studies of prolactin and cortisol effects support the assumption of a direct hormonal action upon IL-12+ monocytes (Petrovsky & Harrison, 1998; Visser et al., 1998; Petrovsky, 2001; Elenkov & Chrousos, 2002; Lange et al., 2006). A direct effect of growth hormone on the immunocompetent cells is less well documented (Elenkov & Chrousos, 2002; Lange et al., 2006). Melatonin also stimulates Type 1 activity (Petrovsky, 2001), and nocturnal suppression of melatonin may counteract this shift.

Irwin et al. (2006) studied in 30 diurnally active healthy adult men (n = 17) and women (n = 13) the monocyte intracellular pro-inflammatory cytokine production across 3 days of baseline testing, and after 1 day of partial sleep deprivation with wakefulness from 23:00 to 03:00. Sampling occurred at 08:00, 12:00, 16:00, 20:00 and 23:00. In the morning after sleep loss, but not at the other times of sampling, the monocyte production of IL-6 and TNFα was significantly greater when compared to the same time (08:00) following uninterrupted sleep. Sleep loss apparently led to an activation of these pro-inflammatory cytokine genes with a more than 3-fold increase in transcription of IL-6


mRNA, and a 2-fold increase in TNFα mRNA. This change was the expression of a functional difference in the monocytes and did not relate to any difference in the numbers of cells. Global gene expression profiling in leukocyte total RNA by high density oligonucleotide assay in five subjects before and after sleep deprivation revealed a set of 22 genes that were significantly upregulated after partial sleep deprivation. These included the circadian clock gene Per1, several epidermal-growth-factor-related genes, and multiple inflammatory response genes. The complex ensemble of functional genomic alterations induced by sleep loss included multiple immediate early response genes, and signal transduction mediators. The remodelling of leukocyte gene expression by sleep and its alteration by sleep loss may point to molecular sites of action in the immune system, and also more generally in cellular physiology and pathology.

(b) Sleep deprivation and the neuroendocrine system

(i) Prolactin and sleep Prolactin plasma concentrations show pulsatile episodic hormone secretion patterns

superimposed upon ultradian rhythms as well as circadian oscillation. The prolactin 24-hour profile reflects both tonic and intermittent hormone release (Veldhuis et al., 1992). The normal secretory pattern of prolactin consists of a series of daily pulses, occurring every 2–3 hours, which vary in amplitude. The bulk of the hormone is secreted during REM sleep. In diurnally active human subjects, REM sleep occurs predominantly during the latter half of the nightly sleep phase, so that the highest plasma prolactin concentrations usually occur late during the night (Sassin et al., 1972, 1973). In men and non-pregnant and non-lactating women, REM sleep is the dominant organizer of prolactin secretion. It has been shown that, in turn, prolactin infusion increases REM activity in the electroencephalogram (Obál et al., 1994; Roky et al., 1995). In lactating women, the reflex elevation of prolactin and oxytocin by nipple stimulation during nursing becomes the predominant controller of circulating prolactin concentrations (Leake et al., 1983).

Sleep onset is associated with an increase in prolactin secretion also during daytime naps, irrespective of the time of the day, but the amplitude of the prolactin rise during daytime sleep is usually less than during nocturnal sleep. Conversely, modest elevations in prolactin concentration may occur at the time of the usual sleep onset even when one remains awake. Thus, prolactin plasma concentrations appear to be regulated by a circadian rhythm and superimposed pulsatile secretions modulated by the sleep-wakefulness pattern, with maximal secretion when sleep and circadian rhythmicity are in phase (Spiegel et al., 1994, 1999; Waldstreicher et al., 1996). Shallow and fragmented sleep, prolonged awakening, and interrupted sleep patterns, as frequently seen in the elderly, are associated with a dampening of the nocturnal prolactin rise, decreased amplitude of the nocturnal prolactin pulses (van Coevorden et al., 1991; Greenspan et al., 1990), and decreased prolactin concentrations (Spiegel et al., 1995).

Prolactine secretion in man is normally restrained by the action of dopamine, which is secreted from the hypothalamus. Prolactin is the only pituitary hormone that is secreted at

SHIFTWORK 713

unrestrained high levels when completely isolated from any tropic influences of the hypothalamus. However, a variety of stimulatory prolactin secretagogues have been identified including steroids (estrogen), hypothalamic peptides, vasoactive intestinal peptide, and oxytocin, and growth factors such as epidermal growth factor, and fibroblast growth factor-2. Numerous medications used in everyday clinical practice elevate prolactin secretion, and this can mask physiological rhythmicity and occasionally may even lead to symptomatic hyperprolactinaemia. These agents include commonly used antidepressants, antiemetics, and narcotics, which antagonize dopamine action or elevate serotonin or endorphin bioactivity (Ben-Jonathan, 1994). Hypnotics like benzodiazepines (e.g. triazolam) and imidazopyridines (e.g. zolpidem) taken at bedtime (concordant with the tendency of the daily prolactin rise) may lead to substantial rises of serum prolactin concentrations into the range regarded as abnormal (Copinschi et al., 1990, 1995). Melatonin itself acutely stimulates prolactin release in humans (Wright et al., 1986; Waldhauser et al., 1987). Endogenous estrogens play a role in the differential regulation of prolactin in relation to age and sex. Mean prolactin concentrations, pulse amplitude, and pulse frequency are all higher in normally cycling young women than in either postmenopausal women or in men (Katznelson et al., 1998). Blunting of the nocturnal rise is not specific and is found also in other medical conditions, including breast cancer.

(ii) Prolactin and the immune system The effects of prolactin in the human body are manyfold. Of importance is the

regulatory role it plays on the immune system. Prolactin receptors are found on most immune-precursor and -effector cells in each of the major haematopoietic and lymphopoietic organs, such as the bone marrow, spleen, and thymus. However, the action of prolactin upon the immune system is complex, and depends upon the stage of both the circadian timing of the prolactin rhythm and its time relations to the circadian rhythms of immune-related functions in the target organs (Cincotta et al., 1995).

In laboratory experiments, prolactin restores immune competence in hypophysectomized animals (Gala, 1991). Inhibition of prolactin secretion by bromocriptine results in immunosuppression (Hiestand et al., 1986; Bernton et al., 1988; Berczi, 1989). Prolactin antagonizes the immunosuppressive effects of glucocorticoids (Bernton et al., 1992). While lowered prolactin concentration leads to immunodeficiency, and exogenous prolactin in short-term experiments produces immunoenhancement, persistently elevated prolactin levels, due to a variety of medical conditions, are associated with immunosuppression (Karmali et al., 1974; Jungers et al., 1982; Gerli et al., 1987; Lavalle et al., 1987; Nicoletti et al., 1989; McMurray et al., 1991; Vidaller et al., 1986, 1992).

Some of the discordant results of investigations pertaining to the effects of prolactin on the immune system may be due to the pronounced circadian variation in its regulatory action (Cincotta et al., 1995), and the marked time-dependent difference of immunocellular responses. In the male BALB/c mouse, the immunostimulatory activity of prolactin was restricted to only an 8-hour daily interval, from 4–12 hours after light on


in animals kept on an LD12:12 regimen. Prolactin administration outside this sensitive interval was occasionally associated with immunosuppressive effects both in the one-way mixed lymphocyte reaction and in the hapten-specific delayed-type hypersensitivity responses. Reducing endogenous levels of prolactin with bromocriptine inhibited immune functions only when the medication was administered during this daily interval of immunoregulatory sensitivity to the hormone (Cincotta et al., 1995). This observation is similar to that of Bernton et al. (1992) who found that the effect of the prolactin inhibitor cysteamine (a dopamine β-hydroxylase inhibitor) on splenocyte mitogenic response was circadian-time-dependent.

A chronobiological explanation of the interaction of the rhythms in human prolactin secretion and in target cell responsiveness, however, has not been reported. It appears that in the immunoregulatory action of prolactin, the overall level of plasma prolactin is of less importance than the circadian rhythmicity of prolactin and that of the apparently circadian periodic responses of the immunocompetent target cell systems. The phase relation between the circadian rhythm in prolactin and that of the rhythms in immunocellular response may be the determining factor for the prolactin effect upon the immune system. Circadian rhythm disruption or phase shifts of either of these rhythms may be associated with immunological dysfunction, which may be of interest for shiftwork and transmeridian flights, and in the elderly, circadian rhythm and sleep disturbances.

Prolactin effects on human immune activity and immunological disorders, including lupus erythematosus and the postpartum exacerbation of rheumatoid arthritis, have been reported (Vidaller et al., 1986; Gerli et al., 1987; Lavalle et al., 1987; Nicoletti et al., 1989; Gala, 1991).

(iii) Hypothalamic–pituitary–thyroid axis and sleep The hypothalamic–pituitary–thyroid (Hth–Pit–Thy) axis possesses an intricate time

structure with rhythmic variations of multiple frequencies found at all levels of the system, from the hypothalamic neurons to the cells of the peripheral target tissues. The frequencies observed include pulsatile secretions and ultradian, circadian, and circannual rhythms. The time-dependent rhythmic (and non-rhythmic) variations of the Hth–Pit–Thy system interact with, and are modulated by, similar time-dependent variations of other neuroendocrine, metabolic, and immune functions.

The thyrotropin-releasing hormone is a tripeptide neurotransmitter that exerts multiple actions in the central nervous system and beyond (Metcalf & Jackson, 1989; Nicolau & Haus, 1992). It is produced also in peripheral tissues, including the immune system (Simard et al., 1989). In addition to its capacity to stimulate the release of TSH from the anterior pituitary, it also stimulates prolactin.

TSH is secreted from the pituitary gland in a series of discrete pulses with an average pulse frequency of 9 pulses/24 hours (range 7–12) in normal men and women (Brabant et al., 1990; Nicolau & Haus, 1992). These pulses are not equally distributed, but cluster during the evening and night hours when fusion of the pulses and an increase in amplitude leads to the nightly increase of TSH concentration, forming the circadian rhythm of this

SHIFTWORK 715

hormone with a maximum in day–night-synchronized subjects occurring usually between 02:00 and 04:00 (Brabant et al., 1990; Samuels et al., 1990). The relatively high peak values of individual TSH pulses during sleep have to be kept in mind, as the values reached may be slightly above the usually accepted normal range.

The pulse pattern may be necessary for normal thyroid gland function due to the better response of the pituitary thyrotrops to intermittent rather than continuous thyrotropin-releasing hormone stimulation (Spencer et al., 1980). Loss of the usual nocturnal variation in TSH and pulse amplitude may be sufficient to cause clinical hypothyroidism (Samuels et al., 1990).

Sleep deprivation and sleep fragmentation result in a marked decrease in the mean 24-hour TSH secretion as well as a lowering of pulse amplitude also without change in peak frequency (Behrends et al., 1998; Brabant et al., 1990; Spiegel et al., 1999).

(iv) Growth hormone and sleep The 24-hour profile of growth hormone (GH) in adult subjects consists of stable low

values interrupted by secretory pulses. There is a marked sexual dimorphism of the secretory pattern. In men, the highest pulse, amounting to about 70% of the secretory output per 24 hours, occurs shortly after sleep onset with the first phase of slow-wave sleep (Van Cauter et al., 1998). In normally cycling women, there is a wider distribution of GH pulses throughout the day. The sleep-onset-associated pulse is still found in most women, but accounts only for a smaller fraction of the total 24-hour secretory product (Ho et al., 1987). The linkage of a major GH pulse to sleep onset leads to an immediate shift in the circadian rhythm in GH with any change of the sleep–wake cycle, e.g. in shiftworkers and after transmeridian travel over several times zones. This linkage also leads to alterations of GH secretion in the case of sleep irregularities (Van Cauter et al., 1998). The mechanism of this association is based on the hypothalamic relationship of the GH releasing hormone to areas of the brain involved in the regulation of sleep (Krueger & Obál, 1993). Inhibition of endogenous GH releasing hormone action by a specific antagonist or by immunoneutralization inhibits both sleep and GH secretion (Ocampo-Lim et al., 1996). On the other hand, substances which promote sleep also lead to increases in nocturnal GH secretion (Gronfier et al., 1996; Van Cauter et al., 1997).

Total sleep deprivation with absence of recovery sleep leads to a markedly decreased growth hormone secretion (Van Cauter et al., 1992; Weibel et al., 1997). Recovery from total sleep deprivation irrespective of the time of day when recovery sleep occurs leads to a robust increase in GH secretion. When bedtime is acutely delayed by a few hours, nocturnal GH levels remain low as long as the subject is awake, and rebound as soon as sleep is initiated (Van Cauter et al., 1998). Semichronic partial sleep deprivation more closely resembles the condition experienced by shiftworkers. Spiegel et al. (2000) studied 11 clinically healthy men after 6 nights of restricted bedtimes (01:00 to 05:00) and after 7 nights of extended bedtimes (21:00 to 09:00). After 1 week of sleep extension to 12 hours, the major GH peak occurred at the same time as the usual 8-hour sleep time after onset of sleep. After 1 week of bedtime reduced to 4 hours, the GH secretory rate


exhibited a biphasic pattern with a large pulse occurring during waking around the usual time of sleep onset on a standard 8-hour bedtime schedule, an expression of the circadian rhythm in GH secretion, followed by a second sleep-induced pulse after onset of the (shortened) sleep span. The state of subchronic partial sleep deprivation (sleep debt) was associated with a markedly different temporal association of GH secretion. The biphasic nature of the GH secretory pattern during sleep restriction resulted in a longer exposure of peripheral tissues to elevated GH concentrations (4 hour 12 minute ± 25 minute vs 3 hour 25 minute ± 33 minute during sleep extension). This biphasic pattern represents an adaptive process to the subchronic sleep deprivation as it was not found in studies with acute sleep deprivation (Van Cauter & Copinschi, 2000). The prolonged exposure of the peripheral tissues to GH may have played a role in the marked deterioration of glucose tolerance that was found in these subjects after 1 week of subchronic sleep restriction (Spiegel et al., 1999). In relation to this study, the question can be raised if a curtailment of sleep by a phase advance due to earlier rising rather than delayed bedtime may avoid the biphasic secretion, and lead to a different adaptive response, possibly with less or different side-effects.

(v) The Hth–Pit–Adr axis and sleep The corticotropic axis with the CRH, the ACTH of the pituitary, and cortisol from the

adrenal cortex is, in addition to the direct effects upon multiple systems, a major messenger of time information in the circadian regulatory system. In addition, CRH is synthesized and produced in multiple peripheral tissues with likely involvement in the regulation of energy balance, metabolism, and immune response (Richard et al., 2000; Baigent, 2001). CRH in the rat brain is produced in the arcuate and paraventricular nuclei of the hypothalamus (Sawchenko & Swanson, 1985), and receives time information from SCN neurons. As a neurotransmitter, CRH acts within the brain to elicit changes in neuroendocrine, autonomic and behavioural activity similar to those observed after stress. Centrally administered CRH induces suppression of NK-cell cytotoxicity (Irwin et al., 1988), an action which appears to be mediated through sympathetic activation as it can be counteracted by adrenergic-receptor blockade (Irwin et al., 1990b). Stress-induced suppression of NK activity appears to be mediated by CRH, and can be antagonized by the central immunoneutralization of CRH (Irwin et al., 1990a). The immunoregulatory role of CRH is not associated with the activation of the pituitary-adrenal axis (Irwin et al., 1990a).

The corticotropic axis receives time information through inputs from oscillator neurons in the SCN to the CRH-ergic neurons in the paraventricular and arcuate nucleus, which release CRH into the hypophyseal portal vein, in a periodic and pulsatile pattern leading to the characteristic periodic and pulsatile ACTH release which is followed by corresponding pulses of cortisol secretion. The clinically manifest activity of the axis reflects the interaction of cycles of hormone secretion and of responsiveness of the endocrine target organs (pituitary and adrenal), and of the corticoid responsive peripheral tissues to the stimulation.

SHIFTWORK 717

The rhythmicity of the corticotropic axis is quite stable and not rapidly altered in its circadian peak by minor changes of the sleep–wakefulness pattern, light, and other environmental stimuli. The normal circadian rhythm of the hypothalamic-pituitary-adrenal axis that is regarded as the major transducer of stress, is primarily regulated by the circadian oscillator system, and is only minimally modulated by sleep. Sleep onset is associated with an acute inhibition of cortisol secretion (Born et al., 1988; Weibel et al., 1995). Awakening during the night, and especially in the morning, is followed by secretory cortisol pulses (Pruessner et al., 1997; Späth-Schwalbe et al., 1991). These changes are absent if a person is prevented from sleeping.

Chronic insomniacs with difficulty falling or staying asleep, with less than 6.5 hours sleep time and a sleep efficacy of less than 80%, exhibited a significantly higher 24-hour ACTH and cortisol secretion than a matched control population with greatest differences in plasma concetrations in the evening and during the first half of the night (Vgontzas et al., 2001). The circadian rhythm in ACTH and cortisol as such was maintained.

In clinically healthy diurnally active young men, partial sleep deprivation (sleep from 04:00 to 08:00) or total sleep deprivation for one night led on the following day to an elevation of plasma cortisol concentration during the evening (18:00 to 23:00), and the onset of the daily quiescent period in plasma cortisol was delayed (Leproult et al., 1997).

In a study of semichronic sleep deprivation in 11 young men whose sleep time was restricted to 4 hours per night (01:00 to 05:00) for 6 nights, Spiegel et al. (1999) found similar changes in the 24-hour profile of plasma cortisol in comparison to the circadian profile of the same subjects studied after a 6-day recovery period. The changes observed after semichronic sleep deprivation also consisted of a shortened quiescent period (537 ± 44 minute versus 634 ± 24 minute) due largely to a delay in its onset of nearly 1.5 hours and raised cortisol concentrations in the afternoon and early evening (Spiegel et al., 1999).

Some studies of sleep loss did not find evidence of a stress reaction in urinary cortisol and catecholamine excretion (Kant et al., 1984) or plasma cortisol (Akerstedt et al., 1980; Davidson et al., 1991; Follenius et al., 1992; Lange et al., 2006; Vgontzas et al., 2004), which, in part, may be due to the relatively short time span in which a deviation from the usual cortisol concentrations can be recognized. These studies suggest that sleep loss does not constitute an acute stimulus for the Hth-Pit-Adr axis, i.e. a “stressor.”

The Hth-Pit-Adr axis possesses powerful and far reaching immunoregulatory activity. CRH directing the characteristic rhythmicity of this system also inhibits endotoxin-stimulated production of IL-1 and IL-6 by human monocytes. ACTH suppresses IFNγ production by human lymphocytes (Johnson et al., 1984).

The glucocorticoids exert an extensive and multifaceted immunoregulatory activity. They are powerful anti-inflammatory agents inhibiting inflammatory mediators including cytokines, phospholipid products, proteases, and oxygen metabolites. They downregulate cytokine expression of IL-1, Il-2, IL-3, IL-6, IL-4, IL-8, IFNγ, and TNFα (for a review, see Petrovsky, 2001). In contrast to the downregulation of cell-mediated immunity, glucocorticoids enhance immunoglobulin production (Cooper et al., 1979) and also


induce the macrophage migration inhibitory factor, a pro-inflammatory cytokine which counteracts and moderates the anti-inflammatory effects of glucocorticoids (Calandra et al., 1995), maintaining a balance between the pro-inflammatory and anti-inflammatory components of the system. By stimulating the production of IL-4, IL-10 and IL-13, glucocorticoids favour the Th2 mode of the immune system (Ramírez et al., 1996). The changes in the immune system found in sleep deprivation and shiftwork may in part be related to the circadian rhythm alterations in the Hth–Pit–Adr system experienced during these conditions.

Sleep loss, similar to aging, may slow down the rate of recovery of the corticotropic axis response following a challenge, and may facilitate the development of central and peripheral disturbances associated with glucosteroid excess. Especially elevated cortisol concentrations at the time of the normal daily quiet period may, in the long run, result in undesirable side-effects, such as memory deficits, insulin resistance, and osteoporosis (Dallman et al., 1993; McEwen, 1998; Dennison et al., 1999; Plat et al., 1999).

In circadian phase shift, as may be experienced in shiftwork or under the effect of competing synchronizers between the light-directed SCN and peripheral stimuli like the time of food uptake, the rhythmic reaction of glucocorticoids inhibits the uncoupling of peripheral circadian oscillators from the central pacemaker (Le Minh et al., 2001). This may counteract the internal circadian desynchronization and favour maintenance of the circadian time organization, and, if a phase shift takes place, determine in part the time of phase adaptation.

(c) Sleep deprivation and metabolism Numerous studies over the last decade have consistently reported, with cross-

sectional as well as with prospective design, an inverse relation between the numbers of hours of sleep and body weight in both children and adults with some age and gender differences noted (Vioque et al., 2000; Sekine et al., 2002; Cournot et al., 2004; Hasler et al., 2004; Patel et al., 2004; Taheri et al., 2004; Reilly et al., 2005). Obesity in shiftworkers has been associated with a short duration of sleep (van Amelsvoort et al., 1999; Moreno et al., 2006). The incidence and degree were related to the duration of the shiftwork. A significant increase in the waist:hips ratio was found in workers after 2–5 years’ involvement in shiftwork, and in the body mass index after more than 5 years in shiftwork (van Amelsvoort et al., 1999). A causal relationship between sleep restriction and weight gain is supported by metabolic studies.

Spiegel et al. (1999, 2004b) studied 24-hour hormone and metabolic profiles in 11 young adult men after 6 days of sleep deprivation (4 hours’ bedtime, 01:00 to 05:00) and 1 week of recovery. After 6 days of sleep deprivation, the mean circadian leptin concentration and the circadian amplitude of leptin were decreased, and ghrelin concentrations were increased together with increased hunger and appetite (Spiegel et al., 2004a, b). The leptin concentrations were similar to the values found after calorie restriction (Chin-Chance et al., 2000) in spite of adequate calorie intake. Similar findings of short sleep duration associated with reduced leptin, elevated ghrelin sampled at a single

SHIFTWORK 719

time point in the morning and increased body mass index was reported by Taheri et al. (2004). In animal experiments, a marked increase in food uptake was found in sleep-deprived rats (Rechtschaffen & Bergmann, 1995). It appears that sleep deprivation alters the regulation of leptin and ghrelin production, and, accordingly, the feedback on the energetic needs and appetite control to the brain, which may lead to an increase in food uptake and represent a risk factor for obesity.

Estimations of the sympathovagal balance derived from recordings of heart-rate variability were significantly higher during sleep restriction (Spiegel et al., 1999). The higher sympathetic activity may be related to metabolic changes (e.g., insulin-resistance) and other metabolic and cardiovascular changes. In sleep deprivation and during sleep-debt conditions, an impairment of carbohydrate tolerance develops with a slower rate of glucose clearance, with a decrease in glucose effectiveness, and a lower acute insulin response to glucose (Spiegel et al., 1999) leading to conditions found in natural aging (Kahn et al., 1993), and close to findings in non-insulin-dependent diabetics (Bergman, 1989) or gestational diabetes (Catalano et al., 1993). Decreased carbohydrate tolerance and increased sympathetic tone are risk factors for the development of insulin resistance, obesity, and hypertension (Reaven et al., 1996), corresponding to the condition described as “metabolic syndrome.” It appears likely that some endemic disorders of the modern society like diabetes and obesity are in part a consequence of chronic sleep deprivation (Sekine et al., 2002; Taheri et al., 2004; Cizza et al., 2005; Gangwisch et al., 2005; Taheri, 2006). This includes the increased incidence of obesity in shiftworkers (van Amelsvoort et al., 1999; Gangwisch et al., 2005; Moreno et al., 2006), which again, may be related to an increased cancer risk in these workers.

The high intake of dietary fat at night by rotating shiftworkers (40% of total calories) (Lennernäs et al., 1994) leads to marked postprandial increases in triacylglycerols and non-esterified fatty acids such as linoleic acid (Holmbäck et al., 2002). Linoleic acid provides a robust stimulatory signal for cancer growth via its mutagenic metabolite, 13-hydroxyoctadecadienoic acid (13-HODE). Elevated physiological nocturnal melatonin levels in the blood of human premenopausal women have the capacity to inhibit the uptake of linoleic acid, and its metabolism to 13-HODE, and tumour proliferative activity in (estrogen receptor negative/progesterone receptor negative (ER–/PgR–) and ER+/PgR+ tissue in isolated breast cancer xenografts perfused in situ. Exposure of these subjects to bright white light at night suppresses melatonin production resulting in substantially increased linoleic acid uptake, 13-HODE formation, and tumour proliferative activity in human breast cancer xenografts perfused in situ, with this melatonin-depleted blood. These results suggest that nocturnal circadian melatonin levels in women may protect against the breast cancer growth-promoting effects of increased dietary linoleic acid levels ingested at night (Blask et al., 2005b).

4.3 Mechanistic arguments

Melatonin has been shown to have antiproliferative effects on human cancer cells cultured in vitro. These oncostatic effects have been observed at physiological


concentrations, and include reduction of cell-cycle progression by increasing the expression of the tumour suppressor gene TP53, and inhibition of DNA synthesis. In addition, melatonin reduces the invasive and metastatic properties of human cancer cells in vitro, and increases intercellular communication between these cells. There is evidence from animal models that melatonin inhibits or reduces the induction of DNA damage by free radicals. Pinealectomized rats showed a higher level of DNA damage in response to treatment with a carcinogen than did pineal-intact rats. Melatonin has also been shown to upregulate anti-oxidant enzyme systems.

Epidemiological studies on genetic polymorphisms in clock-related genes and phenotypes such as morning/evening preference and depressive symptoms, have shown a significant association between a single-nucleotide polymorphism in the PER2 gene and diurnal preference. In a wider sense, the circadian clock may function as a tumour suppressor at the systemic, cellular, and molecular levels. Clock-controlled genes involved in cell-cycle control include c-MYC, MDM2, TP53 and GADD45a, as well as caspases, cyclins, and various transcription factors. In transgenic mice, a deletion in Per2 results in a shorter circadian period, a higher susceptibility to radiation-induced tumours, and reduced apoptosis in thymocytes. The disruption of the circadian rhythm in mice is associated with the accelerated growth of malignant tumours.

Functional loss of the Period genes has been observed in various human tumours, and is probably based on epigenetic changes, i.e. the modulation of the methylation pattern in the promoter region. The loss of the clock protein function and the aberrant methylation of PER1, PER2, PER3, CRY1 and CRY2 promoters has been found in tumours of the breast, endometrium, lung, and in leukaemia. Artificially induced expression of PER1 in non-small lung cancer cells in vitro results in a significant reduction in growth.

The human circadian gene PER3 is linked to breast cancer risk. A polymorphic repeat region in this gene results in a PER3 protein of different length, which is associated with delayed sleep-phase syndrome, and diurnal preference. The variant genotypes are associated with an increased breast cancer risk in premenopausal women.

4.4 References

Abe M, Herzog ED, Yamazaki S et al. (2002). Circadian rhythms in isolated brain regions. J Neurosci, 22:350–356. PMID:11756518

Akerstedt T, Palmblad J, de la Torre B et al. (1980). Adrenocortical and gonadal steroids during sleep deprivation. Sleep, 3:23–30. PMID:6781027

Akhtar RA, Reddy AB, Maywood ES et al. (2002). Circadian cycling of the mouse liver transcriptome, as revealed by cDNA microarray, is driven by the suprachiasmatic nucleus. Curr Biol, 12:540–550 doi:10.1016/S0960-9822(02)00759-5. PMID:11937022

Albrecht U, Sun ZS, Eichele G, Lee CC (1997). A differential response of two putative mammalian circadian regulators, mper1 and mper2, to light. Cell, 91:1055–1064 doi:10.1016/S0092-8674(00)80495-X. PMID:9428527

SHIFTWORK 721

Amalric R, Gautherie M, Hobbins WB et al. (1981). [The future of women with isolated abnormal infrared thermogram of the breast (author’s transl)]. Nouv Presse Med, 10:3153–3155. PMID:7290978

Anderson TD, Arceco R, Hayes TJ (1993). Comparative toxicity and pathology associated with administration of recombinant HuIL-1 alpha to animals. Int Rev Exp Pathol, 34 Pt A;9–36. PMID:8454419

Anderson TJ, Ferguson DJP, Raab GM (1982). Cell turnover in the “resting” human breast: influence of parity, contraceptive pill, age and laterality. Br J Cancer, 46:376–382. PMID:7126427

Archer SN, Robilliard DL, Skene DJ et al. (2003). A length polymorphism in the circadian clock gene Per3 is linked to delayed sleep phase syndrome and extreme diurnal preference. Sleep, 26:413–415. PMID:12841365

Arendt J (1978). Melatonin assays in body fluids. J Neural Transm Suppl, (13):265–278. PMID: 288853

Arendt J (1985). Mammalian pineal rhythms. Pineal Res Rev, 3:161–213. Arendt J (1986). Role of the pineal gland and melatonin in seasonal reproductive function in

mammals. Oxf Rev Reprod Biol, 8:266–320. PMID:3540805 Arendt J (1988). Melatonin. Clin Endocrinol (Oxf), 29:205–229 doi:10.1111/j.1365-

2265.1988.tb00263.x. PMID:3073883 Arendt J, Labib MH, Bojkowski C et al. (1989). Rapid decrease in melatonin production during

treatment of delayed puberty with oestradiol in a case of craniopharyngioma. Lancet, 333:1326. doi:10.1016/S0140-6736(89)92716-5

Arendt J (1995). Melatonin and the Mammalian Pineal Gland. Chapman & Hall: London. Arendt J (2000). Melatonin, circadian rhythms, and sleep. N Engl J Med, 343:1114–1116

doi:10.1056/NEJM200010123431510. PMID:11027748 Arendt J (2005). Melatonin: characteristics, concerns, and prospects. J Biol Rhythms, 20:291–303

doi:10.1177/0748730405277492. PMID:16077149 Arendt J, Borbely AA, Franey C, Wright J (1984). The effects of chronic, small doses of melatonin

given in the late afternoon on fatigue in man: a preliminary study. Neurosci Lett, 45:317–321 doi:10.1016/0304-3940(84)90245-3. PMID:6728321

Arendt J, Labib MH, Bojkowski C et al. (1989). Rapid decrease in melatonin production during treatment of delayed puberty with oestradiol in a case of craniopharyngioma. Lancet, 333:1326. doi:10.1016/S0140-6736(89)92716-5

Axelrod J, Weissbach H (1960). Enzymatic O-methylation of N-acetylserotonin to melatonin. Science, 131:1312 doi:10.1126/science.131.3409.1312. PMID:13795316

Baigent SM (2001). Peripheral corticotropin-releasing hormone and urocortin in the control of the immune response. Peptides, 22:809–820 doi:10.1016/S0196-9781(01)00395-3. PMID:11337095

Baker FC, Driver HS (2007). Circadian rhythms, sleep, and the menstrual cycle. Sleep Med, 8:613–622 doi:10.1016/j.sleep.2006.09.011. PMID:17383933

Baldwin WS, Barrett JC (1998). Melatonin attenuates hydrogen peroxide toxicity in MCF7 cells only at pharmacological concentrations. Biochem Biophys Res Commun, 250:602–605 doi:10.1006/bbrc.1998.9370. PMID:9784392

Baldwin WS, Travlos GS, Risinger JI, Barrett JC (1998). Melatonin does not inhibit estradiol-stimulated proliferation in MCF-7 and BG-1 cells. Carcinogenesis, 19:1895–1900 doi:10.1093/carcin/19.11.1895. PMID:9854999


Ballard PL, Baxter JD, Higgins SJ et al. (1974). General presence of glucocorticoid receptors in mammalian tissues. Endocrinology, 94:998–1002 doi:10.1210/endo-94-4-998. PMID:4362047

Balsalobre A, Brown SA, Marcacci L et al. (2000). Resetting of circadian time in peripheral tissues by glucocorticoid signaling. Science, 289:2344–2347 doi:10.1126/science.289.5488.2344. PMID:11009419

Balsalobre A, Damiola F, Schibler U (1998). A serum shock induces circadian gene expression in mammalian tissue culture cells. Cell, 93:929–937 doi:10.1016/S0092-8674(00)81199-X. PMID:9635423

Bargiello TA, Jackson FR, Young MW (1984). Restoration of circadian behavioural rhythms by gene transfer in Drosophila. Nature, 312:752–754 doi:10.1038/312752a0. PMID:6440029

Barnes RG, Deacon SJ, Forbes MJ, Arendt J (1998a). Adaptation of the 6-sulphatoxymelatonin rhythm in shiftworkers on offshore oil installations during a 2-week 12-h night shift. Neurosci Lett, 241:9–12 doi:10.1016/S0304-3940(97)00965-8. PMID:9502203

Barnes RG, Forbes MJ, Arendt J (1998b). Shift type and season affect adaptation of the 6-sulphatoxymelatonin rhythm in offshore oil rig workers. Neurosci Lett, 252:179–182 doi:10.1016/S0304-3940(98)00585-0. PMID:9739990

Bartness TJ, Song CK, Demas GE (2001). SCN efferents to peripheral tissues: implications for biological rhythms. J Biol Rhythms, 16:196–204. PMID:11407779

Bartsch C, Bartsch H (2006). The anti-tumor activity of pineal melatonin and cancer enhancing life styles in industrialized societies. Cancer Causes Control, 17:559–571 doi:10.1007/s10552-005-9011-8. PMID:16596311

Bartsch H, Bartsch C (1981). Effect of melatonin on experimental tumors under different photoperiods and times of administration. J Neural Transm, 52:269–279 doi:10.1007/BF01256752. PMID:7334363

Bartsch H, Buchberger A, Franz H et al. (2000). Effect of melatonin and pineal extracts on human ovarian and mammary tumor cells in a chemosensitivity assay. Life Sci, 67:2953–2960 doi:10.1016/S0024-3205(00)00882-1. PMID:11133007

Beck-Friis J, Ljunggren JG, Thorén M et al. (1985). Melatonin, cortisol and ACTH in patients with major depressive disorder and healthy humans with special reference to the outcome of the dexamethasone suppression test. Psychoneuroendocrinology, 10:173–186 doi:10.1016/0306-4530(85)90055-1. PMID:2994141

Beck-Friis J, von Rosen D, Kjellman BF et al. (1984). Melatonin in relation to body measures, sex, age, season and the use of drugs in patients with major affective disorders and healthy subjects. Psychoneuroendocrinology, 9:261–277 doi:10.1016/0306-4530(84)90005-2. PMID:6494381

Behrends J, Prank K, Dogu E, Brabant G (1998). Central nervous system control of thyrotropin secretion during sleep and wakefulness. Horm Res, 49:173–177 doi:10.1159/000023167. PMID:9550121

Bellipanni G, Bianchi P, Pierpaoli W et al. (2001). Effects of melatonin in perimenopausal and menopausal women: a randomized and placebo controlled study. Exp Gerontol, 36:297–310 doi:10.1016/S0531-5565(00)00217-5. PMID:11226744

Ben-Jonathan N (1994). Regulation of prolactin secretion. In: Imura H, Ed.. The Pituitary Gland. 2nd ed. New York: Raven Press. pp. 261.

Berczi I (1989). Immunoregulation by neuroendocrine factors. Dev Comp Immunol, 13:329–341 doi:10.1016/0145-305X(89)90042-6. PMID:2572464

SHIFTWORK 723

Berga SL, Mortola JF, Girton L et al. (1989). Neuroendocrine aberrations in women with functional hypothalamic amenorrhea. J Clin Endocrinol Metab, 68:301–308 doi:10.1210/jcem-68-2-301. PMID:2493024

Berga SL, Mortola JF, Yen SS (1988). Amplification of nocturnal melatonin secretion in women with functional hypothalamic amenorrhea. J Clin Endocrinol Metab, 66:242–244 doi:10.1210/jcem-66-1-242. PMID:3335608

Bergman RN (1989). Lilly lecture 1989. Toward physiological understanding of glucose tolerance. Minimal-model approach. Diabetes, 38:1512–1527 doi:10.2337/diabetes.38.12.1512. PMID:2684710

Bernton E, Bryant H, Holaday J, Dave J (1992). Prolactin and prolactin secretagogues reverse immunosuppression in mice treated with cysteamine, glucocorticoids, or cyclosporin-A. Brain Behav Immun, 6:394–408 doi:10.1016/0889-1591(92)90038-P. PMID:1336994

Bernton EW, Meltzer MS, Holaday JW (1988). Suppression of macrophage activation and T-lymphocyte function in hypoprolactinemic mice. Science, 239:401–404 doi:10.1126/science.3122324. PMID:3122324

Berson DM, Dunn FA, Takao M (2002). Phototransduction by retinal ganglion cells that set the circadian clock. Science, 295:1070–1073 doi:10.1126/science.1067262. PMID:11834835

Beskonakli E, Palaoglu S, Aksaray S et al. (2001). Effect of pinealectomy on immune parameters in rats with Staphylococcus aureus infection. Neurosurg Rev, 24:26–30 doi:10.1007/PL00011962. PMID:11339464

Bittman EL, Dempsey RJ, Karsch FJ (1983). Pineal melatonin secretion drives the reproductive response to daylength in the ewe. Endocrinology, 113:2276–2283 doi:10.1210/endo-113-6-2276. PMID:6641634

Bjarnason GA, Jordan RC, Sothern RB (1999). Circadian variation in the expression of cell-cycle proteins in human oral epithelium. Am J Pathol, 154:613–622. PMID:10027418

Bjarnason GA, Jordan RC, Wood PA et al. (2001). Circadian expression of clock genes in human oral mucosa and skin: association with specific cell-cycle phases. Am J Pathol, 158:1793–1801. PMID:11337377

Blask DE (1993). Melathonin in oncology. In: Yu HS & Reiter RJ, Ed. elatonin: biosynthesis physiological effects and clinical applications. Boca Raton: CRC press. pp. 447–475.

Blask DE (2001) An overview of the neuroendocrine regulation of experimental tumor growth by melatonin and its analogues and the therapeutic use of melatonin in oncology. In: Bartsch C, Bartsch H & Blask DE, Eds. The Pineal Gland and Cancer: Neuroimmunoendocrine Mechanisms in Malignancy. Berlin: Springer. pp. 309–342.

Blask DE, Brainard GC, Dauchy RT et al. (2005b). Melatonin-depleted blood from premenopausal women exposed to light at night stimulates growth of human breast cancer xenografts in nude rats. Cancer Res, 65:11174–11184 doi:10.1158/0008-5472.CAN-05-1945. PMID:16322268

Blask DE, Dauchy RT, Sauer LA (2005a). Putting cancer to sleep at night: the neuroendocrine/circadian melatonin signal. Endocrine, 27:179–188 doi:10.1385/ENDO:27:2:179. PMID:16217131

Blask DE, Hill SM (1986). Effects of melatonin on cancer: studies on MCF-7 human breast cancer cells in culture. J Neural Transm, Suppl 21:433–449. PMID: 3462341

Blask DE, Sauer LA, Dauchy RT (2002). Melatonin as a chronobiotic/anticancer agent: cellular, biochemical, and molecular mechanisms of action and their implications for circadian-based


cancer therapy. Curr Top Med Chem, 2:113–132 doi:10.2174/1568026023394407. PMID:11899096

Blask DE, Wilson ST, Zalatan F (1997). Physiological melatonin inhibition of human breast cancer cell growth in vitro: evidence for a glutathione-mediated pathway. Cancer Res, 57:1909–1914. PMID:9157984

Boivin DB, Duffy JF, Kronauer RE, Czeisler CA (1996). Dose-response relationships for resetting of human circadian clock by light. Nature, 379:540–542 doi:10.1038/379540a0. PMID:8596632

Bojkowski CJ, Aldhous ME, English J et al. (1987a). Suppression of nocturnal plasma melatonin and 6-sulphatoxymelatonin by bright and dim light in man. Horm Metab Res, 19:437–440 doi:10.1055/s-2007-1011846. PMID:3692439

Bojkowski CJ, Arendt J (1988). Annual changes in 6-sulphatoxymelatonin excretion in man. Acta Endocrinol (Copenh), 117:470–476. PMID:3389039

Bojkowski CJ, Arendt J, Shih MC, Markey SP (1987b). Melatonin secretion in humans assessed by measuring its metabolite, 6-sulfatoxymelatonin. Clin Chem, 33:1343–1348. PMID:3608151

Born J, Muth S, Fehm HL (1988). The significance of sleep onset and slow wave sleep for nocturnal release of growth hormone (GH) and cortisol. Psychoneuroendocrinology, 13:233–243 doi:10.1016/0306-4530(88)90021-2. PMID:3406323

Brabant G, Prank K, Ranft U et al. (1990). Physiological regulation of circadian and pulsatile thyrotropin secretion in normal man and woman. J Clin Endocrinol Metab, 70:403–409 doi:10.1210/jcem-70-2-403. PMID:2105332

Brainard GC, Hanifin JP, Greeson JM et al. (2001). Action spectrum for melatonin regulation in humans: evidence for a novel circadian photoreceptor. J Neurosci, 21:6405–6412. PMID:11487664

Brismar K, Werner S, Thorén M, Wetterberg L (1985). Metyrapone: an agent for melatonin as well as ACTH and cortisol secretion. J Endocrinol Invest, 8:91–95. PMID:2993405

Broadway J, Arendt J, Folkard S (1987). Bright light phase shifts the human melatonin rhythm during the Antarctic winter. Neurosci Lett, 79:185–189 doi:10.1016/0304-3940(87)90694-X. PMID:3670728

Brown R, Pang G, Husband AJ, King MG (1989b). Suppression of immunity to influenza virus infection in the respiratory tract following sleep disturbance. Reg Immunol, 2:321–325. PMID:2562046

Brown R, Price RJ, King MG, Husband AJ (1989a). Interleukin-1 beta and muramyl dipeptide can prevent decreased antibody response associated with sleep deprivation. Brain Behav Immun, 3:320–330 doi:10.1016/0889-1591(89)90031-7. PMID:2575411

Brun J, Claustrat B, David M (1987). Urinary melatonin, LH, oestradiol, progesterone excretion during the menstrual cycle or in women taking oral contraceptives. Acta Endocrinol (Copenh), 116:145–149. PMID:3661054

Bryant PA, Trinder J, Curtis N (2004). Sick and tired: Does sleep have a vital role in the immune system? Nat Rev Immunol, 4:457–467 doi:10.1038/nri1369. PMID:15173834

Brzezinski A, Fibich T, Cohen M et al. (1992). Effects of melatonin on progesterone production by human granulosa lutein cells in culture. Fertil Steril, 58:526–529. PMID:1521647

Brzezinski A, Lynch HJ, Seibel MM et al. (1988). The circadian rhythm of plasma melatonin during the normal menstrual cycle and in amenorrheic women. J Clin Endocrinol Metab, 66:891–895 doi:10.1210/jcem-66-5-891. PMID:3129448

SHIFTWORK 725

Brzezinski A, Vangel MG, Wurtman RJ et al. (2005). Effects of exogenous melatonin on sleep: a meta-analysis. Sleep Med Rev, 9:41–50 doi:10.1016/j.smrv.2004.06.004. PMID:15649737

Buijs RM, Kalsbeek A (2001). Hypothalamic integration of central and peripheral clocks. Nat Rev Neurosci, 2:521–526 doi:10.1038/35081582. PMID:11433377

Buijs RM, Wortel J, Van Heerikhuize JJ et al. (1999). Anatomical and functional demonstration of a multisynaptic suprachiasmatic nucleus adrenal (cortex) pathway. Eur J Neurosci, 11:1535–1544 doi:10.1046/j.1460-9568.1999.00575.x. PMID:10215906

Bunger MK, Wilsbacher LD, Moran SM et al. (2000). Mop3 is an essential component of the master circadian pacemaker in mammals. Cell, 103:1009–1017 doi:10.1016/S0092-8674(00)00205-1. PMID:11163178

Burch JB, Yost MG, Johnson W, Allen E (2005). Melatonin, sleep, and shift work adaptation. J Occup Environ Med, 47:893–901 doi:10.1097/01.jom.0000177336.21147.9f. PMID:16155474

Burgess HJ, Sharkey KM, Eastman CI (2002). Bright light, dark and melatonin can promote circadian adaptation in night shift workers. Sleep Med Rev, 6:407–420. PMID:12531129

Cagnacci A, Elliott JA, Yen SS (1992). Melatonin: a major regulator of the circadian rhythm of core temperature in humans. J Clin Endocrinol Metab, 75:447–452 doi:10.1210/jc.75.2.447. PMID:1639946

Cagnacci A, Paoletti AM, Soldani R et al. (1995a). Melatonin enhances the luteinizing hormone and follicle-stimulating hormone responses to gonadotropin-releasing hormone in the follicular, but not in the luteal, menstrual phase. J Clin Endocrinol Metab, 80:1095–1099 doi:10.1210/jc.80.4.1095. PMID:7714075

Cagnacci A, Soldani R, Yen SS (1995b). Melatonin enhances cortisol levels in aged but not young women. Eur J Endocrinol, 133:691–695 doi:10.1530/eje.0.1330691. PMID:8548054

Cajochen C, Jud C, Münch M et al. (2006). Evening exposure to blue light stimulates the expression of the clock gene PER2 in humans. Eur J Neurosci, 23:1082–1086 doi:10.1111/j.1460-9568.2006.04613.x. PMID:16519674

Cajochen C, Zeitzer JM, Czeisler CA, Dijk D-J (2000). Dose-response relationship for light intensity and ocular and electroencephalographic correlates of human alertness. Behav Brain Res, 115:75–83 doi:10.1016/S0166-4328(00)00236-9. PMID:10996410

Calandra T, Bernhagen J, Metz CN et al. (1995). MIF as a glucocorticoid-induced modulator of cytokine production. Nature, 377:68–71 doi:10.1038/377068a0. PMID:7659164

Cardinali DP, Lynch HJ, Wurtman RJ (1972). Binding of melatonin to human and rat plasma proteins. Endocrinology, 91:1213–1218 doi:10.1210/endo-91-5-1213. PMID:4538504

Caroleo MC, Nisticò G, Doria G (1992). Effect of melatonin on the immune system. Pharmacol Res, 26 Suppl 2;34–37 doi:10.1016/1043-6618(92)90587-2. PMID:1409318

Carrillo-Vico A, Calvo JR, Abreu P et al. (2004). Evidence of melatonin synthesis by human lymphocytes and its physiological significance: possible role as intracrine, autocrine, and/or paracrine substance. FASEB J, 18:537–539. PMID: 14715696

Carrillo-Vico A, Guerrero JM, Lardone PJ, Reiter RJ (2005). A review of the multiple actions of melatonin on the immune system. Endocrine, 27:189–200 doi:10.1385/ENDO:27:2:189. PMID:16217132

Carter DS, Hall VD, Tamarkin L, Goldman BD (1982). Pineal is required for testicular maintenance in the turkish hamster (mesocricetus brandti). Endocrinology, 111:863–871 doi:10.1210/endo-111-3-863. PMID:6809448


Catalano PM, Tyzbir ED, Wolfe RR et al. (1993). Carbohydrate metabolism during pregnancy in control subjects and women with gestational diabetes. Am J Physiol, 264:E60–E67. PMID:8430789

Challet E, Pévet P (2003). Interactions between photic and nonphotic stimuli to synchronize the master circadian clock in mammals. Front Biosci, 8:s246–s257 doi:10.2741/1039. PMID:12700025

Chen ST, Choo KB, Hou MF et al. (2005). Deregulated expression of the PER1, PER2 and PER3 genes in breast cancers. Carcinogenesis, 26:1241–1246 doi:10.1093/carcin/bgi075. PMID:15790588

Cheng HY, Papp JW, Varlamova O et al. (2007). microRNA modulation of circadian-clock period and entrainment. Neuron, 54:813–829 doi:10.1016/j.neuron.2007.05.017. PMID:17553428

Cheng MY, Bullock CM, Li C et al. (2002). Prokineticin 2 transmits the behavioural circadian rhythm of the suprachiasmatic nucleus. Nature, 417:405–410 doi:10.1038/417405a. PMID:12024206

Chin-Chance C, Polonsky KS, Schoeller DA (2000). Twenty-four-hour leptin levels respond to cumulative short-term energy imbalance and predict subsequent intake. J Clin Endocrinol Metab, 85:2685–2691 doi:10.1210/jc.85.8.2685. PMID:10946866

Cho K (2001). Chronic ‘jet lag’ produces temporal lobe atrophy and spatial cognitive deficits. Nat Neurosci, 4:567–568 doi:10.1038/88384. PMID:11369936

Cho K, Ennaceur A, Cole JC, Suh CK (2000). Chronic jet lag produces cognitive deficits. J Neurosci, 20:RC66. PMID:10704520

Choi NW, Howe GR, Miller AB et al. (1978). An epidemiologic study of breast cancer. Am J Epidemiol, 107:510–521. PMID:665665

Cincotta AH, Knisely TL, Landry RJ et al. (1995). The immunoregulatory effects of prolactin in mice are time of day dependent. Endocrinology, 136:2163–2171 doi:10.1210/en.136.5.2163. PMID:7720666

Cini G, Neri B, Pacini A et al. (2005). Antiproliferative activity of melatonin by transcriptional inhibition of cyclin D1 expression: a molecular basis for melatonin-induced oncostatic effects. J Pineal Res, 39:12–20 doi:10.1111/j.1600-079X.2004.00206.x. PMID:15978052

Cizza G, Skarulis M, Mignot E (2005). A link between short sleep and obesity: building the evidence for causation. Sleep, 28:1217–1220. PMID:16295203

Claustrat B, Brun J, Chazot G (2005). The basic physiology and pathophysiology of melatonin. Sleep Med Rev, 9:11–24 doi:10.1016/j.smrv.2004.08.001. PMID:15649735

Cohen M, Lippman M, Chabner B (1978). Role of pineal gland in aetiology and treatment of breast cancer. Lancet, 312:814–816 doi:10.1016/S0140-6736(78)92591-6. PMID:81365

Colombo LL, Chen GJ, Lopez MC, Watson RR (1992). Melatonin induced increase in gamma-interferon production by murine splenocytes. Immunol Lett, 33:123–126 doi:10.1016/0165-2478(92)90035-M. PMID:1446916

Cooper DA, Duckett M, Petts V, Penny R (1979). Corticosteroid enhancement of immunoglobulin synthesis by pokeweed mitogen-stimulated human lymphocytes. Clin Exp Immunol, 37:145–151. PMID:487651

Copinschi G, Akseki E, Moreno-Reyes R et al. (1995). Effects of bedtime administration of zolpidem on circadian and sleep-related hormonal profiles in normal women. Sleep, 18:417–424. PMID:7481412

SHIFTWORK 727

Copinschi G, Van Onderbergen A, L’Hermite-Balériaux M et al. (1990). Effects of the short-acting benzodiazepine triazolam, taken at bedtime, on circadian and sleep-related hormonal profiles in normal men. Sleep, 13:232–244. PMID:2356395

Cos S, Fernández R (2000). Melatonin effects on intercellular junctional communication in MCF-7 human breast cancer cells. J Pineal Res, 29:166–171 doi:10.1034/j.1600-079X.2000.290306.x. PMID:11034114

Cos S, Fernández R, Güézmes A, Sánchez-Barceló EJ (1998). Influence of melatonin on invasive and metastatic properties of MCF-7 human breast cancer cells. Cancer Res, 58:4383–4390. PMID:9766668

Cos S, González A, Güezmes A et al. (2006). Melatonin inhibits the growth of DMBA-induced mammary tumors by decreasing the local biosynthesis of estrogens through the modulation of aromatase activity. Int J Cancer, 118:274–278 doi:10.1002/ijc.21401. PMID:16080194

Cos S, Martínez-Campa C, Mediavilla MD, Sánchez-Barceló EJ (2005). Melatonin modulates aromatase activity in MCF-7 human breast cancer cells. J Pineal Res, 38:136–142 doi:10.1111/j.1600-079X.2004.00186.x. PMID:15683469

Cos S, Mediavilla MD, Fernández R et al. (2002). Does melatonin induce apoptosis in MCF-7 human breast cancer cells in vitro? J Pineal Res, 32:90–96 doi:10.1034/j.1600-079x.2002.1821.x. PMID:12071473

Cos S, Sánchez-Barceló EJ (1995). Melatonin inhibition of MCF-7 human breast-cancer cells growth: influence of cell proliferation rate. Cancer Lett, 93:207–212 doi:10.1016/0304-3835(95)03811-A. PMID:7621430

Cournot M, Ruidavets JB, Marquié JC et al. (2004). Environmental factors associated with body mass index in a population of Southern France. Eur J Cardiovasc Prev Rehabil, 11:291–297. PMID:15292762 doi:10.1097/01.hjr.0000129738.22970.62

Currier NL, Sun LZ-Y, Miller SC (2000). Exogenous melatonin: quantitative enhancement in vivo of cells mediating non-specific immunity. J Neuroimmunol, 104:101–108 doi:10.1016/S0165-5728(99)00271-4. PMID:10713348

Czeisler CA, Allan JS, Strogatz SH et al. (1986). Bright light resets the human circadian pacemaker independent of the timing of the sleep-wake cycle. Science, 233:667–671 doi:10.1126/science.3726555. PMID:3726555

Czeisler CA, Klerman EB (1999). Circadian and sleep-dependent regulation of hormone release in humans. Recent Prog Horm Res, 54:97–130, discussion 130–132. PMID:10548874

Czeisler CA, Wright KP Jr (1999). Neurobiology of Sleep and Circadian Rhythms. Marcel Decker, Inc: New York.

Dallman MF, Strack AM, Akana SF et al. (1993). Feast and famine: critical role of glucocorticoids with insulin in daily energy flow. Front Neuroendocrinol, 14:303–347 doi:10.1006/frne.1993.1010. PMID:8258378

Davidson JR, Moldofsky H, Lue FA (1991). Growth hormone and cortisol secretion in relation to sleep and wakefulness. J Psychiatry Neurosci, 16:96–102. PMID:1911740

Davis FC (1997). Melatonin: role in development. J Biol Rhythms, 12:498–508 doi:10.1177/074873049701200603. PMID:9406023

Dawson D, Lack L, Morris M (1993). Phase resetting of the human circadian pacemaker with use of a single pulse of bright light. Chronobiol Int, 10:94–102 doi:10.3109/07420529309059697. PMID:8500194


Deacon S, Arendt J (1996). Adapting to phase shifts, I. An experimental model for jet lag and shift work. Physiol Behav, 59:665–673 doi:10.1016/0031-9384(95)02147-7. PMID:8778850

Deacon SJ, Arendt J (1995). Melatonin-induced temperature suppression and its acute phase-shifting effects correlate in a dose-dependent manner in humans. Brain Res, 688:77–85 doi:10.1016/0006-8993(95)96872-I. PMID:8542325

del Gobbo V, Libri V, Villani N et al. (1989). Pinealectomy inhibits interleukin-2 production and natural killer activity in mice. Int J Immunopharmacol, 11:567–573 doi:10.1016/0192-0561(89)90187-2. PMID:2807631

del Río B, García Pedrero JM, Martínez-Campa C et al. (2004). Melatonin, an endogenous-specific inhibitor of estrogen receptor alpha via calmodulin. J Biol Chem, 279:38294–38302 doi:10.1074/jbc.M403140200. PMID:15229223

Demas GE, Nelson RJ (1998). Exogenous melatonin enhances cell-mediated, but not humoral, immune function in adult male deer mice (Peromyscus maniculatus). J Biol Rhythms, 13:245–252 doi:10.1177/074873098129000084. PMID:9615288

Demisch L, Demisch K, Nickelsen T (1988). Influence of dexamethasone on nocturnal melatonin production in healthy adult subjects. J Pineal Res, 5:317–322 doi:10.1111/j.1600-079X.1988.tb00657.x. PMID:3404401

Dennison E, Hindmarsh P, Fall C et al. (1999). Profiles of endogenous circulating cortisol and bone mineral density in healthy elderly men. J Clin Endocrinol Metab, 84:3058–3063 doi:10.1210/jc.84.9.3058. PMID:10487665

Dewan EM (1967). On the possibility of a perfect rhythm method of birth control by periodic light stimulation. Am J Obstet Gynecol, 99:1016–1019. PMID:6058735

Di W-L, Kadva A, Djahanbakhch O, Silman R (1998). Radioimmunoassay of bound and free melatonin in plasma. Clin Chem, 44:304–310. PMID:9474029

Díaz E, Fernández C, Castrillón PO et al. (1999). Effect of exogenous melatonin on neuroendocrine-reproductive function of middle-aged female rats. J Reprod Fertil, 117:331–337 doi:10.1530/jrf.0.1170331. PMID:10690201

Diaz E, Pazo D, Esquifino AI, Diaz B (2000). Effects of ageing and exogenous melatonin on pituitary responsiveness to GnRH in rats. J Reprod Fertil, 119:151–156 doi:10.1530/reprod/119.1.151. PMID:10864825

Díaz López B, Díaz Rodríguez E, Urquijo C, Fernández Alvarez C (2005). Melatonin influences on the neuroendocrine-reproductive axis. Ann N Y Acad Sci, 1057:337–364 doi:10.1196/annals.1356.026. PMID:16399905

Díaz Rodríguez E, Díaz López B, Debeljuk L et al. (1999). Developmental changes of hypothalamic, pituitary and striatal tachykinins in response to testosterone: influence of prenatal melatonin. Peptides, 20:501–508 doi:10.1016/S0196-9781(99)00032-7. PMID:10458521

Dimitrov S, Lange T, Fehm HL, Born J (2004a). A regulatory role of prolactin, growth hormone, and corticosteroids for human T-cell production of cytokines. Brain Behav Immun, 18:368–374 doi:10.1016/j.bbi.2003.09.014. PMID:15157954

Dimitrov S, Lange T, Tieken S et al. (2004b). Sleep associated regulation of T helper 1/T helper 2 cytokine balance in humans. Brain Behav Immun, 18:341–348 doi:10.1016/j.bbi.2003.08.004. PMID:15157951

Ding JH, Sheckter CB, Drinkwater BL et al. (1988). High serum cortisol levels in exercise-associated amenorrhea. Ann Intern Med, 108:530–534. PMID:3348561

SHIFTWORK 729

Dinges DF, Douglas SD, Zaugg L et al. (1994). Leukocytosis and natural killer cell function parallel neurobehavioral fatigue induced by 64 hours of sleep deprivation. J Clin Invest, 93:1930–1939 doi:10.1172/JCI117184. PMID:7910171

Djeridane Y, Touitou Y (2004). Ex vivo studies on the acute and chronic effects of DHEA and DHEA-sulfate on melatonin synthesis in young- and old-rat pineal glands. Steroids, 69:343–349 doi:10.1016/j.steroids.2004.03.003. PMID:15219412

Drazen DL, Bilu D, Bilbo SD, Nelson RJ (2001). Melatonin enhancement of splenocyte proliferation is attenuated by luzindole, a melatonin receptor antagonist. Am J Physiol Regul Integr Comp Physiol, 280:R1476–R1482. PMID:11294771

Dubocovich ML (1995). Melatonin receptors: are there multiple subtypes? Trends Pharmacol Sci, 16:50–56 doi:10.1016/S0165-6147(00)88978-6. PMID:7762083

Dubocovich ML, Markowska M (2005). Functional MT1 and MT2 melatonin receptors in mammals. Endocrine, 27:101–110 doi:10.1385/ENDO:27:2:101. PMID:16217123

Dubocovich ML, Yun K, Al-Ghoul WM et al. (1998). Selective MT2 melatonin receptor antagonists block melatonin-mediated phase advances of circadian rhythms. FASEB J, 12:1211–1220. PMID:9737724

Duffield GE, Best JD, Meurers BH et al. (2002). Circadian programs of transcriptional activation, signaling, and protein turnover revealed by microarray analysis of mammalian cells. Curr Biol, 12:551–557 doi:10.1016/S0960-9822(02)00765-0. PMID:11937023

Duffy JF, Dijk DJ, Hall EF, Czeisler CA (1999). Relationship of endogenous circadian melatonin and temperature rhythms to self-reported preference for morning or evening activity in young and older people. J Investig Med, 47:141–150. PMID:10198570

Duffy JF, Wright KP Jr (2005). Entrainment of the human circadian system by light. J Biol Rhythms, 20:326–338 doi:10.1177/0748730405277983. PMID:16077152

Duffy JF, Zeitzer JM, Rimmer DW et al. (2002). Peak of circadian melatonin rhythm occurs later within the sleep of older subjects. Am J Physiol Endocrinol Metab, 282:E297–E303. PMID:11788360

Dumont M, Benhaberou-Brun D, Paquet J (2001). Profile of 24-h light exposure and circadian phase of melatonin secretion in night workers. J Biol Rhythms, 16:502–511 doi:10.1177/074873001129002178. PMID:11669423

Eastman CI, Boulos Z, Terman M et al. (1995). Light treatment for sleep disorders: consensus report. VI. Shift work. J Biol Rhythms, 10:157–164 doi:10.1177/074873049501000208. PMID:7632989

Eastman CI, Martin SK (1999). How to use light and dark to produce circadian adaptation to night shift work. Ann Med, 31:87–98 doi:10.3109/07853899908998783. PMID:10344580

Ebisawa T, Uchiyama M, Kajimura N et al. (2001). Association of structural polymorphisms in the human period3 gene with delayed sleep phase syndrome. EMBO Rep, 2:342–346 doi:10.1093/embo-reports/kve070. PMID:11306557

Elenkov IJ, Chrousos GP (2002). Stress hormones, proinflammatory and antiinflammatory cytokines, and autoimmunity. Ann N Y Acad Sci, 966:290–303 doi:10.1111/j.1749-6632.2002.tb04229.x. PMID:12114286

Everson CA (1993). Sustained sleep deprivation impairs host defense. Am J Physiol, 265:R1148–R1154. PMID:8238617

Farewell VT, Math B, Math M (1977). The combined effect of breast cancer risk factors. Cancer, 40:931–936 doi:10.1002/1097-0142(197708)40:2<931::AID-CNCR2820400251>3.0.CO;2-Y. PMID:890675


Fernández B, Díaz E, Colmenero MD, Díaz B (1995). Maternal pineal gland participates in prepubertal rats’ ovarian oocyte development. Anat Rec, 243:461–465 doi:10.1002/ar.1092430408. PMID:8597292

Fernández B, Malde JL, Montero A, Acuña D (1990). Relationship between adenohypophyseal and steroid hormones and variations in serum and urinary melatonin levels during the ovarian cycle, perimenopause and menopause in healthy women. J Steroid Biochem, 35:257–262 doi:10.1016/0022-4731(90)90282-W. PMID:2308340

Fernández Alvarez C, Díaz Rodriguez E, Pazo Vinuesa D et al. (1999). In vitro pituitary responsiveness to LHRH in young and old female rats. Influence of melatonin. Mech Ageing Dev, 112:75–83 doi:10.1016/S0047-6374(99)00080-9. PMID:10656184

Ferrari E, Magri F, Dori D et al. (1995). Neuroendocrine correlates of the aging brain in humans. Neuroendocrinology, 61:464–470 doi:10.1159/000126869. PMID:7783860

Ferreira ZS, Fernandes PACM, Duma D et al. (2005). Corticosterone modulates noradrenaline-induced melatonin synthesis through inhibition of nuclear factor kappa B. J Pineal Res, 38:182–188 doi:10.1111/j.1600-079X.2004.00191.x. PMID:15725340

Fève-Montange M, Abou-Samra AB (1983) Glucocorticoids inhibit the in vivo melatonin production by rat pineal stimulated by norepinephrine (NE) or 8-bromo-cyclic AMP. 14th Acta Endocrinol. Congress Satellite Symposium, Abstract 11.

Fève-Montange M, Van Cauter E, Refetoff S et al. (1981). Effects of “jet lag” on hormonal patterns. II. Adaptation of melatonin circadian periodicity. J Clin Endocrinol Metab, 52:642–649 doi:10.1210/jcem-52-4-642. PMID:7204537

Fideleff H, Aparicio NJ, Guitelman A et al. (1976). Effect of melatonin on the basal and stimulated gonadotropin levels in normal men and postmenopausal women. J Clin Endocrinol Metab, 42:1014–1017 doi:10.1210/jcem-42-6-1014. PMID:777019

Filipovich AH, Mathur A, Kamat D, Shapiro RS (1992). Primary immunodeficiencies: genetic risk factors for lymphoma. Cancer Res, 52 Suppl;5465s–5467s. PMID:1327508

Filipski E, King VM, Li X et al. (2002). Host circadian clock as a control point in tumor progression. J Natl Cancer Inst, 94:690–697. PMID:11983758

Folkard S, Akerstedt T (2004). Trends in the risk of accidents and injuries and their implications for models of fatigue and performance. Aviat Space Environ Med, 75 Suppl:A161–A167. PMID:15018280

Folkard S, Lombardi DA (2004). Toward a “Risk Index” to assess work schedules. Chronobiol Int, 21:1063–1072 doi:10.1081/CBI-200036919. PMID:15646251

Follenius M, Brandenberger G, Bandesapt JJ et al. (1992). Nocturnal cortisol release in relation to sleep structure. Sleep, 15:21–27. PMID:1557591

Foster RG (1998). Shedding light on the biological clock. Neuron, 20:829–832 doi:10.1016/S0896-6273(00)80464-X. PMID:9620688

Foster RG, Hankins MW, Peirson SN (2007). Light, photoreceptors, and circadian clocks. [Review]. Methods Mol Biol, 362:3–28 doi:10.1007/978-1-59745-257-1_1. PMID:17416998

Foster RG, Provencio I, Hudson D et al. (1991). Circadian photoreception in the retinally degenerate mouse (rd/rd). J Comp Physiol A Neuroethol Sens Neural Behav Physiol, 169:39–50 doi:10.1007/BF00198171. PMID: 1941717

Freedman MS, Lucas RJ, Soni B et al. (1999). Regulation of mammalian circadian behavior by non-rod, non-cone, ocular photoreceptors. Science, 284:502–504 doi:10.1126/science.284.5413.502. PMID:10205061

SHIFTWORK 731

Fu L, Lee CC (2003). The circadian clock: pacemaker and tumour suppressor. Nat Rev Cancer, 3:350–361 doi:10.1038/nrc1072. PMID:12724733

Fu L, Pelicano H, Liu J et al. (2002). The circadian gene Period2 plays an important role in tumor suppression and DNA damage response in vivo. Cell, 111:41–50 doi:10.1016/S0092-8674(02)00961-3. PMID:12372299

Funk D, Amir S (1999). Conditioned fear attenuates light-induced suppression of melatonin release in rats. Physiol Behav, 67:623–626 doi:10.1016/S0031-9384(99)00073-6. PMID:10549903

Gagne D, Pons M, Philibert D (1985). RU 38486: a potent antiglucocorticoid in vitro and in vivo. J Steroid Biochem, 23:247–251 doi:10.1016/0022-4731(85)90401-7. PMID:2864478

Gail MH, Brinton LA, Byar DP et al. (1989). Projecting individualized probabilities of developing breast cancer for white females who are being examined annually. J Natl Cancer Inst, 81:1879–1886 doi:10.1093/jnci/81.24.1879. PMID:2593165

Gala RR (1991). Prolactin and growth hormone in the regulation of the immune system. Proc Soc Exp Biol Med, 198:513–527. PMID:1891468

Gangwisch JE, Malaspina D, Boden-Albala B, Heymsfield SB (2005). Inadequate sleep as a risk factor for obesity: analyses of the NHANES I. Sleep, 28:1289–1296. PMID:16295214

Gao GC, Dashwood MR, Wei ET (1991). Corticotropin-releasing factor inhibition of substance P-induced vascular leakage in rats: possible sites of action. Peptides, 12:639–644 doi:10.1016/0196-9781(91)90113-4. PMID:1717957

Gautherie M (1983). Thermobiological assessment of benign and malignant breast diseases. Am J Obstet Gynecol, 147:861–869. PMID:6650622

Gautherie M, Gros C (1977). Circadian rhythm alteration of skin temperature in breast cancer. Chronobiologia, 4:1–17. PMID:880849

Gautherie M, Gros CM (1980). Breast thermography and cancer risk prediction. Cancer, 45:51–56 doi:10.1002/1097-0142(19800101)45:1<51::AID-CNCR2820450110>3.0.CO;2-L. PMID:7351006

Geissmann F, Jung S, Littman DR (2003). Blood monocytes consist of two principal subsets with distinct migratory properties. Immunity, 19:71–82 doi:10.1016/S1074-7613(03)00174-2. PMID:12871640

Gerli R, Riccardi C, Nicoletti I et al. (1987). Phenotypic and functional abnormalities of T lymphocytes in pathological hyperprolactinemia. J Clin Immunol, 7:463–470 doi:10.1007/BF00915056. PMID:2961788

Gery S, Komatsu N, Baldjyan L et al. (2006). The circadian gene per1 plays an important role in cell growth and DNA damage control in human cancer cells. Mol Cell, 22:375–382 doi:10.1016/j.molcel.2006.03.038. PMID:16678109

Gery S, Komatsu N, Kawamata N et al. (2007). Epigenetic silencing of the candidate tumor suppressor gene Per1 in non-small cell lung cancer. Clin Cancer Res, 13:1399–1404 doi:10.1158/1078-0432.CCR-06-1730. PMID:17332281

Gibbs M, Hampton S, Morgan L, Arendt J (2002). Adaptation of the circadian rhythm of 6-sulphatoxymelatonin to a shift schedule of seven nights followed by seven days in offshore oil installation workers. Neurosci Lett, 325:91–94 doi:10.1016/S0304-3940(02)00247-1. PMID:12044629

Gibbs M, Hampton S, Morgan L, Arendt J (2007). Predicting circadian response to abrupt phase shift: 6-sulphatoxymelatonin rhythms in rotating shift workers offshore. J Biol Rhythms, 22:368–370 doi:10.1177/0748730407302843. PMID:17660453


Gibertini M, Graham C, Cook MR (1999). Self-report of circadian type reflects the phase of the melatonin rhythm. Biol Psychol, 50:19–33 doi:10.1016/S0301-0511(98)00049-0. PMID:10378437

Gilad E, Zisapel N (1995). High-affinity binding of melatonin to hemoglobin. Biochem Mol Med, 56:115–120 doi:10.1006/bmme.1995.1066. PMID:8825074

Gillette MU, Mitchell JW (2002). Signaling in the suprachiasmatic nucleus: selectively responsive and integrative. Cell Tissue Res, 309:99–107 doi:10.1007/s00441-002-0576-1. PMID:12111540

Giordano M, Vermeulen M, Palermo MS (1993). Seasonal variations in antibody-dependent cellular cytotoxicity regulation by melatonin. FASEB J, 7:1052–1054. PMID:8370475

Glaser R, MacCallum RC, Laskowski BF et al. (2001). Evidence for a shift in the Th-1 to Th-2 cytokine response associated with chronic stress and aging. J Gerontol A Biol Sci Med Sci, 56:M477–M482. PMID: 11487599

Glass JD, Lynch GR (1981). Melatonin: identification of sites of antigonadal action in mouse brain. Science, 214:821–823 doi:10.1126/science.7292016. PMID:7292016

Goldman BD (1999). The circadian timing system and reproduction in mammals. Steroids, 64:679–685 doi:10.1016/S0039-128X(99)00052-5. PMID:10503728

Goldman BD (2001). Mammalian photoperiodic system: formal properties and neuroendocrine mechanisms of photoperiodic time measurement. J Biol Rhythms, 16:283–301 doi:10.1177/074873001129001980. PMID:11506375

González A, Martínez-Campa C, Mediavilla MD et al. (2007). Effects of MT1 melatonin receptor overexpression on the aromatase-suppressive effect of melatonin in MCF-7 human breast cancer cells. Oncol Rep, 17:947–953. PMID:17342341

Graham C, Cook MR, Gerkovich MM, Sastre A (2001). Examination of the melatonin hypothesis in women exposed at night to EMF or bright light. Environ Health Perspect, 109:501–507 doi:10.2307/3454709. PMID:11401762

Greenspan SL, Klibanski A, Rowe JW, Elahi D (1990). Age alters pulsatile prolactin release: influence of dopaminergic inhibition. Am J Physiol, 258:E799–E804. PMID:2333989

Griefahn B, Künemund C, Golka K et al. (2002). Melatonin synthesis: a possible indicator of intolerance to shiftwork. Am J Ind Med, 42:427–436. PMID:12382256 doi:10.1002/ajim.10122

Gronfier C, Luthringer R, Follenius M et al. (1996). A quantitative evaluation of the relationships between growth hormone secretion and delta wave electroencephalographic activity during normal sleep and after enrichment in delta waves. Sleep, 19:817–824. PMID:9085491

Hahn RA (1991). Profound bilateral blindness and the incidence of breast cancer. Epidemiol, 2:208–210 doi:10.1097/00001648-199105000-00008. PMID: 2054403

Hajak G, Rodenbeck A, Ehrenthal HD et al. (1997). No evidence for a physiological coupling between melatonin and glucocorticoids. Psychopharmacology (Berl), 133:313–322 doi:10.1007/s002130050408. PMID:9372529

Haldar C, Rai S, Singh R (2004). Melatonin blocks dexamethasone-induced immunosuppression in a seasonally breeding rodent Indian palm squirrel, Funambulus pennanti. Steroids, 69:367–377 doi:10.1016/j.steroids.2004.03.006. PMID:15219786

Haldar C, Singh R, Guchhait P (2001). Relationship between the annual rhythms in melatonin and immune system status in the tropical palm squirrel, Funambulus pennanti. Chronobiol Int, 18:61–69 doi:10.1081/CBI-100001174. PMID:11247114

SHIFTWORK 733

Hansen MA, Garde HA, Hansen J (2006). Diurnal urinary 6-sulfatoxymelatonin levels among healthy Danish nurses during work and leisure time. Chronobiol Int, 23:1203–1215 doi:10.1080/07420520601100955. PMID:17190706

Hara R, Wan K, Wakamatsu H et al. (2001). Restricted feeding entrains liver clock without participation of the suprachiasmatic nucleus. Genes Cells, 6:269–278 doi:10.1046/j.1365-2443.2001.00419.x. PMID:11260270

Hardeland R (2005). Antioxidative protection by melatonin: multiplicity of mechanisms from radical detoxification to radical avoidance. Endocrine, 27:119–130 doi:10.1385/ENDO:27:2:119. PMID:16217125

Hardeland R, Poeggeler B (2003). Non-vertebrate melatonin. J Pineal Res, 34:233–241 doi:10.1034/j.1600-079X.2003.00040.x. PMID:12662344

Harris W (1977) Fatigue, circadian rhythms, and truck accidents. In: Vigilance Theory, Operational Performance, and Physiological Correlate. New York: Plenum Press. pp. 1033–1046.

Hasler G, Buysse DJ, Klaghofer R et al. (2004). The association between short sleep duration and obesity in young adults: a 13-year prospective study. Sleep, 27:661–666. PMID:15283000

Hastings MH (2000). Circadian clockwork: two loops are better than one. Nat Rev Neurosci, 1:143–146 doi:10.1038/35039080. PMID:11252777

Hastings MH, Herzog ED (2004). Clock genes, oscillators, and cellular networks in the suprachiasmatic nuclei. J Biol Rhythms, 19:400–413 doi:10.1177/0748730404268786. PMID:15534320

Hattar S, Lucas RJ, Mrosovsky N et al. (2003). Melanopsin and rod-cone photoreceptive systems account for all major accessory visual functions in mice. Nature, 424:76–81 doi:10.1038/nature01761. PMID:12808468

Haus E (1964). Periodicity in response and susceptibility to environmental stimuli. Ann N Y Acad Sci, 117:292–319 doi:10.1111/j.1749-6632.1964.tb48187.x. PMID:14196650

Haus E (1996). Biologic rhythms in hematology. Pathol Biol (Paris), 44:618–630. PMID:8977919 Haus E (2007). Chronobiology in the endocrine system. Adv Drug Deliv Rev, 59:985–1014

doi:10.1016/j.addr.2007.01.001. PMID:17804113 Haus E, Halberg F (1969). Phase-shifting of circadian rhythms in rectal temperature, serum

corticosterone and liver glycogen of the male C-mouse. Rass Neurol Veg, 23:83–112. PMID:5398839

Haus E, Lakatua D, Halberg F (1967). The internal timing of several circadian rhythms in the blinded mouse. Exp Med Surg, 25:7–45. PMID:5625877

Haus E, Nicolau GY, Ghinea E et al. (1996). Stimulation of the secretion of dehydroepiandrosterone by melatonin in mouse adrenals in vitro. Life Sci, 58:PL263–PL276 doi:10.1016/0024-3205(96)00079-3. PMID: 8614260

Haus E, Nicolau GY, Lakatua D, Sackett-Lundeen L (1988). Reference values for chronopharmacology. Annu Rev Chronopharm, 4:333–424.

Haus E, Smolensky MH (1999). Biologic rhythms in the immune system. Chronobiol Int, 16:581–622 doi:10.3109/07420529908998730. PMID:10513884

Haus E, Touitou Y (1992) Principles of clinical chronobiology. In: Touitou Y & Haus E, Eds. Biologic Rhythms in Clinical and Laboratory Medicine. Berlin, Heidelberg, Paris: Springer-Verlag. pp. 6–34.

Hazlerigg DG, Andersson H, Johnston JD, Lincoln G (2004). Molecular characterization of the long-day response in the Soay sheep, a seasonal mammal. Curr Biol, 14:334–339. PMID:14972686


Hazlerigg DG, Morgan PJ, Messager S (2001). Decoding photoperiodic time and melatonin in mammals: what can we learn from the pars tuberalis? J Biol Rhythms, 16:326–335 doi:10.1177/074873001129002042. PMID:11506378

Hébert M, Martin SK, Lee C, Eastman CI (2002). The effects of prior light history on the suppression of melatonin by light in humans. J Pineal Res, 33:198–203 doi:10.1034/j.1600-079X.2002.01885.x. PMID:12390501

Herljevic M, Middleton B, Thapan K, Skene DJ (2005). Light-induced melatonin suppression: age-related reduction in response to short wavelength light. Exp Gerontol, 40:237–242 doi:10.1016/j.exger.2004.12.001. PMID:15763401

Hiestand PC, Mekler P, Nordmann R et al. (1986). Prolactin as a modulator of lymphocyte responsiveness provides a possible mechanism of action for cyclosporine. Proc Natl Acad Sci USA, 83:2599–2603 doi:10.1073/pnas.83.8.2599. PMID:2939454

Hildebrandt G, Moser M, Lehofer M (1998). Chronobiology and Chronomedicine - Biologic Rhythms Medical Consequences (German). Stuttgart: Hippokrates.

Hildebrandt G, Rohmert W, Rutenfranz J (1974). 12 and 24 h Rhythms in error frequency of locomotive drivers and the influence of tiredness. Int J Chronobiol, 2:175–180. PMID:4422310

Hill SM, Blask DE (1988). Effects of the pineal hormone melatonin on the proliferation and morphological characteristics of human breast cancer cells (MCF-7) in culture. Cancer Res, 48:6121–6126. PMID:3167858

Hill SM, Spriggs LL, Simon MA et al. (1992). The growth inhibitory action of melatonin on human breast cancer cells is linked to the estrogen response system. Cancer Lett, 64:249–256 doi:10.1016/0304-3835(92)90050-6. PMID:1638517

Ho KY, Evans WS, Blizzard RM et al. (1987). Effects of sex and age on the 24-hour profile of growth hormone secretion in man: importance of endogenous estradiol concentrations. J Clin Endocrinol Metab, 64:51–58 doi:10.1210/jcem-64-1-51. PMID:3782436

Hoffman K (1973). The influence of photoperiod and melatonin on testis size and body weight in the Djungarian hamster. J Comp Physiol, 85:267–282. doi:10.1007/BF00694233

Hogenesch JB, Gu YZ, Jain S, Bradfield CA (1998). The basic-helix-loop-helix-PAS orphan MOP3 forms transcriptionally active complexes with circadian and hypoxia factors. Proc Natl Acad Sci USA, 95:5474–5479 doi:10.1073/pnas.95.10.5474. PMID:9576906

Holmbäck U, Forslund A, Forslund J et al. (2002). Metabolic responses to nocturnal eating in men are affected by sources of dietary energy. J Nutr, 132:1892–1899. PMID:12097665

Honma K, Honma S, Wada T (1987). Phase-dependent shift of free-running human circadian rhythms in response to a single bright light pulse. Experientia, 43:1205–1207 doi:10.1007/BF01945525. PMID:3691737

Honma S, Ikeda M, Abe H et al. (1998). Circadian oscillation of BMAL1, a partner of a mammalian clock gene Clock, in rat suprachiasmatic nucleus. Biochem Biophys Res Commun, 250:83–87 doi:10.1006/bbrc.1998.9275. PMID:9735336

Horne JA, Ostberg O (1976). A self-assessment questionnaire to determine morningness-eveningness in human circadian rhythms. Int J Chronobiol, 4:97–110. PMID:1027738

Howe GR (1984) Epidemiology of radiogenic breast cancer. In: Boice JD& Fraumeni JF, Eds. Radiation Carcinogenesis: Epidemiology and Biological Significance. Raven: New York. pp. 119–129.

Hriscu M, Saulea G, Ostriceanu S, Baciu I (2002). Circadian phagocytic activity in rats under light-dark and constant light regimens. Rom J Physiol, 39-40:17–26. PMID:15984664

SHIFTWORK 735

Hsu DS, Zhao X, Zhao S et al. (1996). Putative human blue-light photoreceptors hCRY1 and hCRY2 are flavoproteins. Biochemistry, 35:13871–13877 doi:10.1021/bi962209o. PMID:8909283

Hua H, Wang Y, Wan C et al. (2006). Circadian gene mPer2 overexpression induces cancer cell apoptosis. Cancer Sci, 97:589–596 doi:10.1111/j.1349-7006.2006.00225.x. PMID:16827798

Iglesias R, Terrés A, Chavarria A (1980). Disorders of the menstrual cycle in airline stewardesses. Aviat Space Environ Med, 51:518–520. PMID:7387577

Illnerová H, Zvolsky P, Vanĕcek J (1985). The circadian rhythm in plasma melatonin concentration of the urbanized man: the effect of summer and winter time. Brain Res, 328:186–189 doi:10.1016/0006-8993(85)91342-3. PMID:3971177

Irwin M, Brown M, Patterson T et al. (1991). Neuropeptide Y and natural killer cell activity: findings in depression and Alzheimer caregiver stress. FASEB J, 5:3100–3107. PMID:1743441

Irwin M, Hauger RL, Brown MR, Britton KT (1988). CRF activates autonomic nervous system and reduces natural killer cytotoxicity. Am J Physiol, 255:R744–R747. PMID:2847561

Irwin M, Hauger RL, Jones L et al. (1990b). Sympathetic nervous system mediates central corticotropin-releasing factor induced suppression of natural killer cytotoxicity. J Pharmacol Exp Ther, 255:101–107. PMID: 2120421

Irwin M, Mascovich A, Gillin JC et al. (1994). Partial sleep deprivation reduces natural killer cell activity in humans. Psychosom Med, 56:493–498. PMID:7871104

Irwin M, McClintick J, Costlow C et al. (1996). Partial night sleep deprivation reduces natural killer and cellular immune responses in humans. FASEB J, 10:643–653. PMID:8621064

Irwin M, Vale W, Rivier C (1990a). Central corticotropin-releasing factor mediates the suppressive effect of stress on natural killer cytotoxicity. Endocrinology, 126:2837–2844 doi:10.1210/endo-126-6-2837. PMID:2161737

Irwin MR, Wang M, Campomayor CO et al. (2006). Sleep deprivation and activation of morning levels of cellular and genomic markers of inflammation. Arch Intern Med, 166:1756–1762 doi:10.1001/archinte.166.16.1756. PMID:16983055

Jakubcakova V, Oster H, Tamanini F et al. (2007). Light entrainment of the mammalian circadian clock by a PRKCA-dependent posttranslational mechanism. Neuron, 54:831–843 doi:10.1016/j.neuron.2007.04.031. PMID:17553429

Janković BD, Knezević Z, Kojić L, Nikolić V (1994). Pineal gland and immune system. Immune functions in the chick embryo pinealectomized at 96 hours of incubation. Ann N Y Acad Sci, 719:398–409. PMID:8010609

Johansson C, Willeit M, Smedh C et al. (2003). Circadian clock-related polymorphisms in seasonal affective disorder and their relevance to diurnal preference. Neuropsychopharmacology, 28:734–739 doi:10.1038/sj.npp.1300121. PMID:12655319

Johnson HM, Torres BA, Smith EM et al. (1984). Regulation of lymphokine (gamma-interferon) production by corticotropin. J Immunol, 132:246–250. PMID:6317743

Johnston JD, Ebling FJ, Hazlerigg DG (2005). Photoperiod regulates multiple gene expression in the suprachiasmatic nuclei and pars tuberalis of the Siberian hamster (Phodopus sungorus). Eur J Neurosci, 21:2967–2974 doi:10.1111/j.1460-9568.2005.04148.x. PMID:15978008

Johnston JD, Messager S, Barrett P, Hazlerigg DG (2003a). Melatonin action in the pituitary: neuroendocrine synchronizer and developmental modulator? J Neuroendocrinol, 15:405–408. PMID:12622841 doi:10.1046/j.1365-2826.2003.00972.x


Johnston JD, Messager S, Ebling FJP et al. (2003b). Gonadotrophin-releasing hormone drives melatonin receptor down-regulation in the developing pituitary gland. Proc Natl Acad Sci USA, 100:2831–2835 doi:10.1073/pnas.0436184100. PMID:12598657

Johnston JD, Schuster C, Barrett P, Hazlerigg DG (2007). Regulation of the ovine MT1 melatonin receptor promoter: interaction between multiple pituitary transcription factors at different phases of development. Mol Cell Endocrinol, 268:59–66 doi:10.1016/j.mce.2007.01.015. PMID:17337323

Johnston JD, Tournier BB, Andersson H et al. (2006). Multiple effects of melatonin on rhythmic clock gene expression in the mammalian pars tuberalis. Endocrinology, 147:959–965 doi:10.1210/en.2005-1100. PMID:16269454

Joshi BN, Troiani ME, Milin J et al. (1986). Adrenal-mediated depression of N-acetyltransferase activity and melatonin levels in the rat pineal gland. Life Sci, 38:1573–1580 doi:10.1016/0024-3205(86)90496-0. PMID:3702592

Jungers P, Dougados M, Pélissier C et al. (1982). Lupus nephropathy and pregnancy. Report of 104 cases in 36 patients. Arch Intern Med, 142:771–776 doi:10.1001/archinte.142.4.771. PMID:7073417

Kahn SE, Prigeon RL, McCulloch DK et al. (1993). Quantification of the relationship between insulin sensitivity and beta-cell function in human subjects. Evidence for a hyperbolic function. Diabetes, 42:1663–1672 doi:10.2337/diabetes.42.11.1663. PMID:8405710

Kalsbeek A, Buijs RM (2002). Output pathways of the mammalian suprachiasmatic nucleus: coding circadian time by transmitter selection and specific targeting. Cell Tissue Res, 309:109–118 doi:10.1007/s00441-002-0577-0. PMID:12111541

Kant GJ, Genser SG, Thorne DR et al. (1984). Effects of 72 hour sleep deprivation on urinary cortisol and indices of metabolism. Sleep, 7:142–146. PMID:6740058

Karmali RA, Lauder I, Horrobin DF (1974). Letter: Prolactin and the immune response. Lancet, 304:106–107 doi:10.1016/S0140-6736(74)91670-5. PMID:4136979

Katzenberg D, Young T, Finn L et al. (1998). A CLOCK polymorphism associated with human diurnal preference. Sleep, 21:569–576. PMID:9779516

Katznelson L, Riskind PN, Saxe VC, Klibanski A (1998). Prolactin pulsatile characteristics in postmenopausal women. J Clin Endocrinol Metab, 83:761–764 doi:10.1210/jc.83.3.761. PMID:9506722

Kauppila A, Kivelä A, Pakarinen A, Vakkuri O (1987). Inverse seasonal relationship between melatonin and ovarian activity in humans in a region with a strong seasonal contrast in luminosity. J Clin Endocrinol Metab, 65:823–828 doi:10.1210/jcem-65-5-823. PMID:3667880

Kawara S, Mydlarski R, Mamelak AJ et al. (2002). Low-dose ultraviolet B rays alter the mRNA expression of the circadian clock genes in cultured human keratinocytes. J Invest Dermatol, 119:1220–1223 doi:10.1046/j.1523-1747.2002.19619.x. PMID:12485420

Kayumov L, Casper RF, Hawa RJ et al. (2005). Blocking low-wavelength light prevents nocturnal melatonin suppression with no adverse effect on performance during simulated shift work. J Clin Endocrinol Metab, 90:2755–2761 doi:10.1210/jc.2004-2062. PMID:15713707

Kennaway DJ, Lushington K, Dawson D et al. (1999). Urinary 6-sulfatoxymelatonin excretion and aging: new results and a critical review of the literature. J Pineal Res, 27:210–220 doi:10.1111/j.1600-079X.1999.tb00617.x. PMID:10551768

SHIFTWORK 737

Khalsa SBS, Jewett ME, Cajochen C, Czeisler CA (2003). A phase response curve to single bright light pulses in human subjects. J Physiol, 549:945–952 doi:10.1113/jphysiol.2003.040477. PMID:12717008

Kiang JG, Poree L, Wei ET (1987). Anti-inflammatory activity of corticotropin releasing factor: II. Mechanisms of action. Proc West Pharmacol Soc, 30:63–65. PMID:3498170

Kiefer T, Ram PT, Yuan L, Hill SM (2002). Melatonin inhibits estrogen receptor transactivation and cAMP levels in breast cancer cells. Breast Cancer Res Treat, 71:37–45 doi:10.1023/A:1013301408464. PMID:11859872

King DP, Zhao Y, Sangoram AM et al. (1997). Positional cloning of the mouse circadian clock gene. Cell, 89:641–653 doi:10.1016/S0092-8674(00)80245-7. PMID:9160755

Kivelä A, Kauppila A, Ylöstalo P et al. (1988). Seasonal, menstrual and circadian secretions of melatonin, gonadotropins and prolactin in women. Acta Physiol Scand, 132:321–327 doi:10.1111/j.1748-1716.1988.tb08335.x. PMID:3147572

Klein DC (2007). Arylalkylamine N-acetyltransferase: “the Timezyme”. J Biol Chem, 282:4233–4237 doi:10.1074/jbc.R600036200. PMID:17164235

Klerman EB, Gershengorn HB, Duffy JF, Kronauer RE (2002). Comparisons of the variability of three markers of the human circadian pacemaker. J Biol Rhythms, 17:181–193 doi:10.1177/074873002129002474. PMID:12002165

Klerman EB, Jayne GA, Smith KA, Czeisler CA (1997). Apparent synchronization of the human circadian pacemaker to a scheduled (T-24 h) cycle of sleep in darkness and wake activity in very dim (20 lux) light in a sighted 22 year old man. Sleep Res, 26:724.

Kornmann B, Preitner N, Rifat D et al. (2001). Analysis of circadian liver gene expression by ADDER, a highly sensitive method for the display of differentially expressed mRNAs. Nucleic Acids Res, 29:E51–E1 doi:10.1093/nar/29.11.e51. PMID:11376163

Kostoglou-Athanassiou I, Treacher DF, Wheeler MJ, Forsling ML (1998). Melatonin administration and pituitary hormone secretion. Clin Endocrinol (Oxf), 48:31–37 doi:10.1046/j.1365-2265.1998.00341.x. PMID:9509065

Kramer A, Yang FC, Snodgrass P et al. (2001). Regulation of daily locomotor activity and sleep by hypothalamic EGF receptor signaling. Science, 294:2511–2515 doi:10.1126/science.1067716. PMID:11752569

Krueger JM, Obál F Jr (1993). Growth hormone-releasing hormone and interleukin-1 in sleep regulation. FASEB J, 7:645–652. PMID:8500689

Krueger JM, Obal F Jr, Opp M et al. (1990a). Somnogenic cytokines and models concerning their effects on sleep. Yale J Biol Med, 63:157–172. PMID:2205056

Krueger JM, Toth LA, Johannsen L, Opp MR (1990b). Infectious disease and sleep: involvement of neuroendocrine-neuroimmune mechanisms. Int J Neurosci, 51:359–362 doi:10.3109/00207459008999744. PMID:2279902

Lakatua DJ, Haus E, Swoyer JK et al. (1975). Meal timing shifts circadian rhythms in serum iron and insulin but not in plasma cortisol in human volunteers. Chronobiologia, 2 Suppl 1;39–40.

Lakatua DJ, Haus M, Berge C et al. (1988). Diet and mealtiming as circadian synchronizers. Annu Rev Chronopharm, 5:303–306.

Lakatua DJ, White M, Sackett-Lundeen LL, Haus E (1983). Change in phase relations of circadian rhythms in cell proliferation induced by time-limited feeding in BALB/c X DBA/2F1 mice bearing a transplantable Harding-Passey tumor. Cancer Res, 43:4068–4072. PMID:6871848


Lange T, Dimitrov S, Fehm H-L et al. (2006). Shift of monocyte function toward cellular immunity during sleep. Arch Intern Med, 166:1695–1700 doi:10.1001/archinte.166.16.1695. PMID:16983046

Laughlin GA, Loucks AB, Yen SS (1991). Marked augmentation of nocturnal melatonin secretion in amenorrheic athletes, but not in cycling athletes: unaltered by opioidergic or dopaminergic blockade. J Clin Endocrinol Metab, 73:1321–1326 doi:10.1210/jcem-73-6-1321. PMID:1955514

Lavalle C, Loyo E, Paniagua R et al. (1987). Correlation study between prolactin and androgens in male patients with systemic lupus erythematosus. J Rheumatol, 14:268–272. PMID:3298648

Lavie P, Weler B (1989). Timing of naps: effects of post-nap sleepiness levels. Electroencephalogr Clin Neurophysiol, 72:218–224 doi:10.1016/0013-4694(89)90246-0. PMID: 2465124

Lawson NO, Wee BE, Blask DE et al. (1992). Melatonin decreases estrogen receptor expression in the medial preoptic area of inbred (LSH/SsLak) golden hamsters. Biol Reprod, 47:1082–1090 doi:10.1095/biolreprod47.6.1082. PMID:1493172

Le Minh N, Damiola F, Tronche F et al. (2001). Glucocorticoid hormones inhibit food-induced phase-shifting of peripheral circadian oscillators. EMBO J, 20:7128–7136 doi:10.1093/emboj/20.24.7128. PMID:11742989

Leake RD, Waters CB, Rubin RT et al. (1983). Oxytocin and prolactin responses in long-term breast-feeding. Obstet Gynecol, 62:565–568. PMID:6684741

Lee CC (2005). The circadian clock and tumor suppression by mammalian period genes. Methods Enzymol, 393:852–861 doi:10.1016/S0076-6879(05)93045-0. PMID:15817328

Lee E, Salic A, Kirschner MW (2001). Physiological regulation of [beta]-catenin stability by Tcf3 and CK1epsilon. J Cell Biol, 154:983–993 doi:10.1083/jcb.200102074. PMID:11524435

Lennernäs M, Akerstedt T, Hambraeus L (1994). Nocturnal eating and serum cholesterol of three-shift workers. Scand J Work Environ Health, 20:401–406. PMID:7701285

Leon-Blanco MM, Guerrero JM, Reiter RJ et al. (2003). Melatonin inhibits telomerase activity in the MCF-7 tumor cell line both in vivo and in vitro. J Pineal Res, 35:204–211 doi:10.1034/j.1600-079X.2003.00077.x. PMID:12932205

Leon-Blanco MM, Guerrero JM, Reiter RJ, Pozo D (2004). RNA expression of human telomerase subunits TR and TERT is differentially affected by melatonin receptor agonists in the MCF-7 tumor cell line. Cancer Lett, 216:73–80 doi:10.1016/j.canlet.2004.05.003. PMID:15500950

Leproult R, Copinschi G, Buxton O, Van Cauter E (1997). Sleep loss results in an elevation of cortisol levels the next evening. Sleep, 20:865–870. PMID:9415946

Lerner AB, Case JD, Takahashi Y et al. (1958). Isolation of melatonin, the pineal gland factor that lightens melanocytes. J Am Chem Soc, 80:2587 doi:10.1021/ja01543a060.

Levine ME, Milliron AN, Duffy LK (1994). Diurnal and seasonal rhythms of melatonin, cortisol and testosterone in interior Alaska. Arctic Med Res, 53:25–34. PMID:8048998

Lewy AJ, Ahmed S, Jackson JML, Sack RL (1992). Melatonin shifts human circadian rhythms according to a phase-response curve. Chronobiol Int, 9:380–392 doi:10.3109/07420529209064550. PMID:1394610

Lewy AJ, Bauer VK, Ahmed S et al. (1998). The human phase response curve (PRC) to melatonin is about 12 hours out of phase with the PRC to light. Chronobiol Int, 15:71–83 doi:10.3109/07420529808998671. PMID:9493716

Lewy AJ, Sack RL, Miller LS, Hoban TM (1987). Antidepressant and circadian phase-shifting effects of light. Science, 235:352–354 doi:10.1126/science.3798117. PMID:3798117

SHIFTWORK 739

Lewy AJ, Wehr TA, Goodwin FK et al. (1980). Light suppresses melatonin secretion in humans. Science, 210:1267–1269 doi:10.1126/science.7434030. PMID:7434030

Lewy H, Haus E, Ashkenazi IE (2007). Possible linkage between the ability to change the period (tau) of the prolactin and cortisol rhythms in women and breast cancer risk. Chronobiol Int, 24:365–381 doi:10.1080/07420520701282349. PMID:17453854

Liebmann PM, Hofer D, Felsner P et al. (1996). Beta-blockade enhances adrenergic immunosuppression in rats via inhibition of melatonin release. J Neuroimmunol, 67:137–142 doi:10.1016/0165-5728(96)00050-1. PMID:8765337

Lin MC, Kripke DF, Parry BL, Berga SL (1990). Night light alters menstrual cycles. Psychiatry Res, 33:135–138 doi:10.1016/0165-1781(90)90067-F. PMID:2243890

Lincoln G, Messenger S, Andersson H, Hazlerigg D (2002). Temporal expression of seven clock genes in the suprachiasmatic nucleus and the pars tuberalis of the sheep: evidence for an internal coincidence timer. Proc Natl Acad Sci USA, 99:13890–13895 doi:10.1073/pnas.212517599. PMID: 12374857

Lincoln GA (2002). Neuroendocrine regulation of seasonal gonadotrophin and prolactin rhythms: lessons from the Soay ram model. Reprod Suppl, 59:131–147. PMID:12698978

Lincoln GA (2006a). Decoding the nightly melatonin signal through circadian clockwork. Mol Cell Endocrinol, 252:69–73 doi:10.1016/j.mce.2006.03.006. PMID:16647195

Lincoln GA (2006b). Melatonin entrainment of circannual rhythms. Chronobiol Int, 23:301–306 doi:10.1080/07420520500464452. PMID:16687303

Lincoln GA, Andersson H, Hazlerigg D (2003). Clock genes and the long-term regulation of prolactin secretion: evidence for a photoperiod/circannual timer in the pars tuberalis. J Neuroendocrinol, 15:390–397 doi:10.1046/j.1365-2826.2003.00990.x. PMID:12622839

Lincoln GA, Clarke IJ (1994). Photoperiodically-induced cycles in the secretion of prolactin in hypothalamo-pituitary disconnected rams: evidence for translation of the melatonin signal in the pituitary gland. J Neuroendocrinol, 6:251–260 doi:10.1111/j.1365-2826.1994.tb00580.x. PMID:7920591

Linthorst AC, Flachskamm C, Hopkins SJ et al. (1997). Long-term intracerebroventricular infusion of corticotropin-releasing hormone alters neuroendocrine, neurochemical, autonomic, behavioral, and cytokine responses to a systemic inflammatory challenge. J Neurosci, 17:4448–4460. PMID:9151762

Lissoni P, Barni S, Tancini G et al. (1994a). Pineal-opioid system interactions in the control of immunoinflammatory responses. Ann N Y Acad Sci, 741 1 Neuroimmunomo;191–196 doi:10.1111/j.1749-6632.1994.tb39659.x. PMID:7825805

Lissoni P, Barni S, Tancini G et al. (1994b). Role of the pineal gland in the control of macrophage functions and its possible implication in cancer: a study of interactions between tumor necrosis factor-alpha and the pineal hormone melatonin. J Biol Regul Homeost Agents, 8:126–129. PMID:7660855

Lissoni P, Chilelli M, Villa S et al. (2003). Five years survival in metastatic non-small cell lung cancer patients treated with chemotherapy alone or chemotherapy and melatonin: a randomized trial. J Pineal Res, 35:12–15 doi:10.1034/j.1600-079X.2003.00032.x. PMID:12823608

Lissoni P, Marelli O, Mauri R et al. (1986). Ultradian chronomodulation by melatonin of a Placebo effect upon human killer cell activity. Chronobiologia, 13:339–343. PMID:3816407


Liu C, Weaver DR, Strogatz SH, Reppert SM (1997). Cellular construction of a circadian clock: period determination in the suprachiasmatic nuclei. Cell, 91:855–860 doi:10.1016/S0092-8674(00)80473-0. PMID:9413994

Liu F, Ng TB, Fung MC (2001). Pineal indoles stimulate the gene expression of immunomodulating cytokines. J Neural Transm, 108:397–405 doi:10.1007/s007020170061. PMID:11475007

Liu X, Uchiyama M, Shibui K et al. (2000). Diurnal preference, sleep habits, circadian sleep propensity and melatonin rhythm in healthy human subjects. Neurosci Lett, 280:199–202 doi:10.1016/S0304-3940(00)00793-X. PMID:10675795

Lockley SW, Brainard GC, Czeisler CA (2003). High sensitivity of the human circadian melatonin rhythm to resetting by short wavelength light. J Clin Endocrinol Metab, 88:4502–4505 doi:10.1210/jc.2003-030570. PMID:12970330

Lockley SW, Skene DJ, Arendt J et al. (1997). Relationship between melatonin rhythms and visual loss in the blind. J Clin Endocrinol Metab, 82:3763–3770 doi:10.1210/jc.82.11.3763. PMID:9360538

Lockley SW, Skene DJ, James K et al. (2000). Melatonin administration can entrain the free-running circadian system of blind subjects. J Endocrinol, 164:R1–R6 doi:10.1677/joe.0.164R001. PMID:10607943

Lopez-Gonzalez MA, Calvo JR, Osuna C, Guerrero JM (1992). Interaction of melatonin with human lymphocytes: evidence for binding sites coupled to potentiation of cyclic AMP stimulated by vasoactive intestinal peptide and activation of cyclic GMP. J Pineal Res, 12:97–104 doi:10.1111/j.1600-079X.1992.tb00034.x. PMID:1324307

Loucks AB, Mortola JF, Girton L, Yen SS (1989). Alterations in the hypothalamic-pituitary-ovarian and the hypothalamic-pituitary-adrenal axes in athletic women. J Clin Endocrinol Metab, 68:402–411 doi:10.1210/jcem-68-2-402. PMID:2537332

Lovenberg W, Jequier E, Sjoerdsma A (1967). Tryptophan hydroxylation: measurement in pineal gland, brainstem, and carcinoid tumor. Science, 155:217–219 doi:10.1126/science.155.3759.217. PMID:6015530

Luboshitzky R, Lavi S, Thuma I, Lavie P (1995). Increased nocturnal melatonin secretion in male patients with hypogonadotropic hypogonadism and delayed puberty. J Clin Endocrinol Metab, 80:2144–2148 doi:10.1210/jc.80.7.2144. PMID:7608268

Luboshitzky R, Lavi S, Thuma I, Lavie P (1996b). Testosterone treatment alters melatonin concentrations in male patients with gonadotropin-releasing hormone deficiency. J Clin Endocrinol Metab, 81:770–774 doi:10.1210/jc.81.2.770. PMID:8636302

Luboshitzky R, Levi M, Shen-Orr Z et al. (2000). Long-term melatonin administration does not alter pituitary-gonadal hormone secretion in normal men. Hum Reprod, 15:60–65 doi:10.1093/humrep/15.1.60. PMID:10611189

Luboshitzky R, Wagner O, Lavi S et al. (1996a). Decreased nocturnal melatonin secretion in patients with Klinefelter’s syndrome. Clin Endocrinol (Oxf), 45:749–754 doi:10.1046/j.1365-2265.1996.8710881.x. PMID:9039342

Lucas RJ, Freedman MS, Lupi D et al. (2001). Identifying the photoreceptive inputs to the mammalian circadian system using transgenic and retinally degenerate mice. Behav Brain Res, 125:97–102 doi:10.1016/S0166-4328(01)00274-1. PMID:11682100

Lucas RJ, Freedman MS, Muñoz M et al. (1999). Regulation of the mammalian pineal by non-rod, non-cone, ocular photoreceptors. Science, 284:505–507 doi:10.1126/science.284.5413.505. PMID:10205062

SHIFTWORK 741

Lucey DR, Clerici M, Shearer GM (1996). Type 1 and type 2 cytokine dysregulation in human infectious, neoplastic, and inflammatory diseases. Clin Microbiol Rev, 9:532–562. PMID:8894351

Lynch HJ, Ho M, Wurtman RJ (1977). The adrenal medulla may mediate the increase in pineal melatonin synthesis induced by stress, but not that caused by exposure to darkness. J Neural Transm, 40:87–97 doi:10.1007/BF01250561. PMID:192851

MacMahon B, Cole P, Brown J (1973). Etiology of human breast cancer: a review. J Natl Cancer Inst, 50:21–42. PMID:4571238

MacMahon B, Cole P, Lin TM et al. (1970). Age at first birth and breast cancer risk. Bull World Health Organ, 43:209–221. PMID:5312521

Madokoro S, Nakagawa H, Misaki K et al. (1997). Nocturnal melatonin profiles before and one year after beginning shift-work. Psychiatry Clin Neurosci, 51:17–22 doi:10.1111/j.1440-1819.1997.tb02360.x. PMID:9076855

Maestroni GJ (2001). The immunotherapeutic potential of melatonin. Expert Opin Investig Drugs, 10:467–476 doi:10.1517/13543784.10.3.467. PMID:11227046

Maestroni GJ, Conti A (1996). Melatonin and the immune-hematopoietic system therapeutic and adverse pharmacological correlates. Neuroimmunomodulation, 3:325–332 doi:10.1159/000097292. PMID:9266542

Maestroni GJ, Conti A, Pierpaoli W (1986). Role of the pineal gland in immunity. Circadian synthesis and release of melatonin modulates the antibody response and antagonizes the immunosuppressive effect of corticosterone. J Neuroimmunol, 13:19–30 doi:10.1016/0165-5728(86)90047-0. PMID:2944914

Maestroni GJ, Conti A, Pierpaoli W (1988a). Role of the pineal gland in immunity. III. Melatonin antagonizes the immunosuppressive effect of acute stress via an opiatergic mechanism. Immunology, 63:465–469. PMID:3350581

Maestroni GJ, Conti A, Pierpaoli W (1988b). Pineal melatonin, its fundamental immunoregulatory role in aging and cancer. Ann N Y Acad Sci, 521 1 Neuroimmunomo;140–148 doi:10.1111/j.1749-6632.1988.tb35272.x. PMID:3377360

Magri F, Locatelli M, Balza G et al. (1997). Changes in endocrine circadian rhythms as markers of physiological and pathological brain aging. Chronobiol Int, 14:385–396 doi:10.3109/07420529709001459. PMID:9262874

Makkison I, Arendt J (1991). Melatonin secretion in humans on two different antarctic bases (68° and 75°S). J Interdiscipl Cycle Res, 22:149–150.

Malpaux B, Migaud M, Tricoire H, Chemineau P (2001). Biology of mammalian photoperiodism and the critical role of the pineal gland and melatonin. J Biol Rhythms, 16:336–347 doi:10.1177/074873001129002051. PMID:11506379

Mansfield CM, Carabasi RA, Wells W, Borman K (1973). Circadian rhythm in the skin temperature of normal and cancerous breasts. Int J Chronobiol, 1:235–243. PMID:4776534

Marte B (2004). Cell division and cancer. Nature, 432:293–295 doi:10.1038/432293a. Martikainen H, Tapanainen J, Vakkuri O et al. (1985). Circannual concentrations of melatonin,

gonadotrophins, prolactin and gonadal steroids in males in a geographical area with a large annual variation in daylight. Acta Endocrinol (Copenh), 109:446–450. PMID:3929512

Martin JE, Klein DC (1976). Melatonin inhibition of the neonatal pituitary response to luteinizing hormone-releasing factor. Science, 191:301–302 doi:10.1126/science.1108199. PMID:1108199


Martin JE, McKellar S, Klein DC (1980). Melatonin inhibition of the in vivo pituitary response to luteinizing hormone-releasing hormone in the neonatal rat. Neuroendocrinology, 31:13–17 doi:10.1159/000123044. PMID:6993981

Martin JE, Sattler C (1982). Selectivity of melatonin pituitary inhibition for luteinizing hormone-releasing hormone. Neuroendocrinology, 34:112–116 doi:10.1159/000123287. PMID:6122168

Martins E Jr, Fernandes LC, Bartol I et al. (1998). The effect of melatonin chronic treatment upon macrophage and lymphocyte metabolism and function in Walker-256 tumour-bearing rats. J Neuroimmunol, 82:81–89 doi:10.1016/S0165-5728(97)00191-4. PMID:9526849

Masson-Pévet M, George D, Kalsbeek A et al. (1994). An attempt to correlate brain areas containing melatonin-binding sites with rhythmic functions: a study in five hibernator species. Cell Tissue Res, 278:97–106. PMID:7954706 doi:10.1007/BF00305781

Matsui DH, Machado-Santelli GM (1997). Alterations in F-actin distribution in cells treated with melatonin. J Pineal Res, 23:169–175 doi:10.1111/j.1600-079X.1997.tb00351.x. PMID:9462848

Matsuo T, Yamaguchi S, Mitsui S et al. (2003). Control mechanism of the circadian clock for timing of cell division in vivo. Science, 302:255–259 doi:10.1126/science.1086271. PMID:12934012

Matthews CD, Guerin MV, Wang X (1991). Human plasma melatonin and urinary 6-sulphatoxy melatonin: studies in natural annual photoperiod and in extended darkness. Clin Endocrinol (Oxf), 35:21–27 doi:10.1111/j.1365-2265.1991.tb03491.x. PMID:1889136

Maywood ES, Bittman EL, Hastings MH (1996). Lesions of the melatonin- and androgen-responsive tissue of the dorsomedial nucleus of the hypothalamus block the gonadal response of male Syrian hamsters to programmed infusions of melatonin. Biol Reprod, 54:470–477 doi:10.1095/biolreprod54.2.470. PMID:8788201

McArthur AJ, Hunt AE, Gillette MU (1997). Melatonin action and signal transduction in the rat suprachiasmatic circadian clock: activation of protein kinase C at dusk and dawn. Endocrinology, 138:627–634 doi:10.1210/en.138.2.627. PMID:9002996

McEwen BS (1998). Protective and damaging effects of stress mediators. N Engl J Med, 338:171–179 doi:10.1056/NEJM199801153380307. PMID:9428819

McIntyre IM, Norman TR, Burrows GD, Armstrong SM (1989). Human melatonin suppression by light is intensity dependent. J Pineal Res, 6:149–156 doi:10.1111/j.1600-079X.1989.tb00412.x. PMID:2915324

McIntyre IM, Norman TR, Burrows GD, Armstrong SM (1990). Melatonin supersensitivity to dim light in seasonal affective disorder. Lancet, 335:488 doi:10.1016/0140-6736(90)90732-K. PMID:1968219

McMurray R, Keisler D, Kanuckel K et al. (1991). Prolactin influences autoimmune disease activity in the female B/W mouse. J Immunol, 147:3780–3787. PMID:1940367

McNamara P, Seo SB, Rudic RD et al. (2001). Regulation of CLOCK and MOP4 by nuclear hormone receptors in the vasculature: a humoral mechanism to reset a peripheral clock. Cell, 105:877–889 doi:10.1016/S0092-8674(01)00401-9. PMID:11439184

Metcalf G, Jackson IMD, editors (1989). Thyrotropin releasing hormone: Biomedical significance. Ann N Y Acad Sci, 553:1–631. PMID: 2497668

Métivier R, Penot G, Hübner MR et al. (2003). Estrogen receptor-alpha directs ordered, cyclical, and combinatorial recruitment of cofactors on a natural target promoter. Cell, 115:751–763 doi:10.1016/S0092-8674(03)00934-6. PMID:14675539

SHIFTWORK 743

Middleton B, Stone BM, Arendt J (2002). Human circadian phase in 12:12 h, 200: <8 lux and 1000: <8 lux light-dark cycles, without scheduled sleep or activity. Neurosci Lett, 329:41–44 doi:10.1016/S0304-3940(02)00574-8. PMID:12161258

Midwinter MJ, Arendt J (1991). Adaptation of the melatonin rhythm in human subjects following night-shift work in Antarctica. Neurosci Lett, 122:195–198 doi:10.1016/0304-3940(91)90856-O. PMID:2027519

Migaud M, Daveau A, Malpaux B (2005). MTNR1A melatonin receptors in the ovine premammillary hypothalamus: day-night variation in the expression of the transcripts. Biol Reprod, 72:393–398 doi:10.1095/biolreprod.104.030064. PMID:15470001

Miyauchi F, Nanjo K, Otsuka K (1991). [Effects of continuous lighting on secretion of melatonin and pituitary hormones in women]. Nippon Sanka Fujinka Gakkai Zasshi, 43:529–534. PMID:1905333

Miyauchi F, Nanjo K, Otsuka K (1992). [Effects of night shift on plasma concentrations of melatonin, LH, FSH and prolactin, and menstrual irregularity]. Sangyo Igaku, 34:545–550. PMID:1460786

Mocchegiani E, Bulian D, Santarelli L et al. (1996). The zinc pool is involved in the immune-reconstituting effect of melatonin in pinealectomized mice. J Pharmacol Exp Ther, 277:1200–1208. PMID:8667179

Moldofsky H (1994). Central nervous system and peripheral immune functions and the sleep-wake system. J Psychiatry Neurosci, 19:368–374. PMID:7803370

Molinero P, Soutto M, Benot S et al. (2000). Melatonin is responsible for the nocturnal increase observed in serum and thymus of thymosin alpha1 and thymulin concentrations: observations in rats and humans. J Neuroimmunol, 103:180–188 doi:10.1016/S0165-5728(99)00237-4. PMID:10696913

Molis TM, Spriggs LL, Hill SM (1994). Modulation of estrogen receptor mRNA expression by melatonin in MCF-7 human breast cancer cells. Mol Endocrinol, 8:1681–1690 doi:10.1210/me.8.12.1681. PMID:7708056

Møller M, Baeres FMM (2002). The anatomy and innervation of the mammalian pineal gland. Cell Tissue Res, 309:139–150 doi:10.1007/s00441-002-0580-5. PMID:12111544

Monk TH (1992) Chronobiology of mental performance. In: Touitou Y & Haus E, Eds. Biologic Rhythms in Clinical and Laboratory Medicine. Berlin, Heidelberg, Paris: Springer-Verlag. pp. 208–213.

Moore CB, Siopes TD (2003). Melatonin enhances cellular and humoral immune responses in the Japanese quail (Coturnix coturnix japonica) via an opiatergic mechanism. Gen Comp Endocrinol, 131:258–263 doi:10.1016/S0016-6480(03)00011-X. PMID:12714007

Moore CB, Siopes TD, Steele CT, Underwood H (2002). Pineal melatonin secretion, but not ocular melatonin secretion, is sufficient to maintain normal immune responses in Japanese quail (Coturnix coturnix japonica). Gen Comp Endocrinol, 126:352–358 doi:10.1016/S0016-6480(02)00011-4. PMID:12093123

Moreno CRC, Louzada FM, Teixeira LR et al. (2006). Short sleep is associated with obesity among truck drivers. Chronobiol Int, 23:1295–1303 doi:10.1080/07420520601089521. PMID:17190714

Morgan PJ (2000). The pars tuberalis: the missing link in the photoperiodic regulation of prolactin secretion? [Review]. J Neuroendocrinol, 12:287–295 doi:10.1046/j.1365-2826.2000.00459.x. PMID:10718925


Morgan PJ, Barrett P, Howell HE, Helliwell R (1994). Melatonin receptors: localization, molecular pharmacology and physiological significance. Neurochem Int, 24:101–146 doi:10.1016/0197-0186(94)90100-7. PMID:8161940

Morgan PJ, Williams LM (1996). The pars tuberalis of the pituitary: a gateway for neuroendocrine output. Rev Reprod, 1:153–161 doi:10.1530/ror.0.0010153. PMID:9414453

Morin D, Simon N, Deprés-Brummer P et al. (1997). Melatonin high-affinity binding to alpha-1-acid glycoprotein in human serum. Pharmacology, 54:271–275 doi:10.1159/000139495. PMID:9380773

Morin PJ (1999). beta-catenin signaling and cancer. Bioessays, 21:1021–1030 doi:10.1002/(SICI)1521-1878(199912)22:1<1021::AID-BIES6>3.0.CO;2-P. PMID:10580987

Mormont MC, Lévi F (1997). Circadian-system alterations during cancer processes: a review. Int J Cancer, 70:241–247. PMID:9009166 doi:10.1002/(SICI)1097-0215(19970117)70:2<241::AID-IJC16>3.0.CO;2-L

Mormont MC, Waterhouse J (2002). Contribution of the rest-activity circadian rhythm to quality of life in cancer patients. Chronobiol Int, 19:313–323 doi:10.1081/CBI-120002606. PMID:11962684

Morrey KM, McLachlan JA, Serkin CD, Bakouche O (1994). Activation of human monocytes by the pineal hormone melatonin. J Immunol, 153:2671–2680. PMID:8077674

Nagano M, Adachi A, Nakahama K et al. (2003). An abrupt shift in the day/night cycle causes desynchrony in the mammalian circadian center. J Neurosci, 23:6141–6151. PMID:12853433

Nakamura W, Yamazaki S, Takasu NN et al. (2005). Differential response of Period 1 expression within the suprachiasmatic nucleus. J Neurosci, 25:5481–5487 doi:10.1523/JNEUROSCI.0889-05.2005. PMID:15944376

Nakano Y, Miura T, Hara I et al. (1982). The effect of shift work on cellular immune function. J Hum Ergol (Tokyo), 11 Suppl;131–137. PMID:6985367

Nathan PJ, Wyndham EL, Burrows GD, Norman TR (2000). The effect of gender on the melatonin suppression by light: a dose response relationship. J Neural Transm, 107:271–279 doi:10.1007/s007020050022. PMID:10821436

Nelson RJ (2004). Seasonal immune function and sickness responses. Trends Immunol, 25:187–192 doi:10.1016/j.it.2004.02.001. PMID:15039045

Nelson RJ, Drazen DL (2000). Melatonin mediates seasonal changes in immune function. Ann N Y Acad Sci, 917:404–415 doi:10.1111/j.1749-6632.2000.tb05405.x. PMID:11268368

Nelson W, Tong YL, Lee JK, Halberg F (1979). Methods for cosinor-rhythmometry. Chronobiologia, 6:305–323. PMID:548245

Nicolau GY, Haus E (1992) Chronobiology of the hypothalamic-pituitary-thyroid axis. In: Touitou Y & Haus E, Eds. Biologic Rhythms in Clinical and Laboratory Medicine, 2nd Ed. New York: Springer-Verlag. pp. 330–347.

Nicoletti I, Gerli R, Orlandi S et al. (1989). Defective natural killer cell activity in puerperal hyperprolactinemia. J Reprod Immunol, 15:113–121 doi:10.1016/0165-0378(89)90031-4. PMID:2788740

Niles LP, Wang J, Shen L et al. (1999). Melatonin receptor mRNA expression in human granulosa cells. Mol Cell Endocrinol, 156:107–110 doi:10.1016/S0303-7207(99)00135-5. PMID:10612428

Nosjean O, Ferro M, Coge F et al. (2000). Identification of the melatonin-binding site MT3 as the quinone reductase 2. J Biol Chem, 275:31311–31317 doi:10.1074/jbc.M005141200. PMID:10913150

SHIFTWORK 745

Nowak R, McMillen IC, Redman J, Short RV (1987). The correlation between serum and salivary melatonin concentrations and urinary 6-hydroxymelatonin sulphate excretion rates: two non-invasive techniques for monitoring human circadian rhythmicity. Clin Endocrinol (Oxf), 27:445–452 doi:10.1111/j.1365-2265.1987.tb01172.x. PMID:3436070

Obál F Jr, Payne L, Kacsoh B et al. (1994). Involvement of prolactin in the REM sleep-promoting activity of systemic vasoactive intestinal peptide (VIP). Brain Res, 645:143–149 doi:10.1016/0006-8993(94)91647-0. PMID:8062077

Ocampo-Lim B, Guo W, DeMott-Friberg R et al. (1996). Nocturnal growth hormone (GH) secretion is eliminated by infusion of GH-releasing hormone antagonist. J Clin Endocrinol Metab, 81:4396–4399 doi:10.1210/jc.81.12.4396. PMID:8954048

Okatani Y, Sagara Y (1994). Amplification of nocturnal melatonin secretion in women with functional secondary amenorrhoea: relation to endogenous oestrogen concentration. Clin Endocrinol (Oxf), 41:763–770 doi:10.1111/j.1365-2265.1994.tb02791.x. PMID:7889612

Opp MR, Kapás L, Toth LA (1992). Cytokine involvement in the regulation of sleep. Proc Soc Exp Biol Med, 201:16–27. PMID:1382300

Owen J, Arendt J (1992). Melatonin suppression in human subjects by bright and dim light in antarctica: time and season-dependent effects. Neurosci Lett, 137:181–184 doi:10.1016/0304-3940(92)90399-R. PMID:1584458

Ozturk L, Pelin Z, Karadeniz D et al. (1999). Effects of 48 hours sleep deprivation on human immune profile. Sleep Res, 2:107–111. PMID: 11382891

Palmblad J, Cantell K, Strander H et al. (1976). Stressor exposure and immunological response in man: interferon-producing capacity and phagocytosis. J Psychosom Res, 20:193–199 doi:10.1016/0022-3999(76)90020-9. PMID:972359

Palmblad J, Petrini B, Wasserman J, Akerstedt T (1979). Lymphocyte and granulocyte reactions during sleep deprivation. Psychosom Med, 41:273–278. PMID:482523

Panda S, Antoch MP, Miller BH et al. (2002). Coordinated transcription of key pathways in the mouse by the circadian clock. Cell, 109:307–320 doi:10.1016/S0092-8674(02)00722-5. PMID:12015981

Panzer A, Lottering ML, Bianchi P et al. (1998). Melatonin has no effect on the growth, morphology or cell cycle of human breast cancer (MCF-7), cervical cancer (HeLa), osteosarcoma (MG-63) or lymphoblastoid (TK6) cells. Cancer Lett, 122:17–23 doi:10.1016/S0304-3835(97)00360-1. PMID:9464486

Patel SR, Ayas NT, Malhotra MR et al. (2004). A prospective study of sleep duration and mortality risk in women. Sleep, 27:440–444. PMID:15164896

Persengiev SP (1992). 2-(125I) iodomelatonin binding sites in rat adrenals: pharmacological characteristics and subcellular distribution. Life Sci, 51:647–651 doi:10.1016/0024-3205(92)90237-J. PMID:1323736

Persengiev SP, Kanchev LN (1991). Melatonin and adrenal cortex steroid production: in vivo and in vitro studies. Folia Histochem Cytobiol, 29:15–18. PMID:1783093

Petrovsky N (2001). Towards a unified model of neuroendocrine-immune interaction. Immunol Cell Biol, 79:350–357 doi:10.1046/j.1440-1711.2001.01029.x. PMID:11488982

Petrovsky N, Harrison LC (1998). The chronobiology of human cytokine production. Int Rev Immunol, 16:635–649 doi:10.3109/08830189809043012. PMID:9646180

Pévet P, Agez L, Bothorel B et al. (2006). Melatonin in the multi-oscillatory mammalian circadian world. Chronobiol Int, 23:39–51 doi:10.1080/07420520500482074. PMID:16687278


Phillips MJ, Wilson DW, Simpson HW et al. (1981). Characterisation of breast skin temperature rhythms of women in relation to menstrual status. Acta Endocrinol (Copenh), 96:350–360. PMID:7211096

Pioli C, Caroleo MC, Nistico G, Doria G (1993). Melatonin increases antigen presentation and amplifies specific and non specific signals for T-cell proliferation. Int J Immunopharmacol, 15:463–468 doi:10.1016/0192-0561(93)90060-C. PMID:8365822

Pittendrigh CS (1967). Circadian systems. I. The driving oscillation and its assay in Drosophila pseudoobscura. Proc Natl Acad Sci USA, 58:1762–1767 doi:10.1073/pnas.58.4.1762. PMID:5237901

Plat L, Leproult R, L’Hermite-Baleriaux M et al. (1999). Metabolic effects of short-term elevations of plasma cortisol are more pronounced in the evening than in the morning. J Clin Endocrinol Metab, 84:3082–3092 doi:10.1210/jc.84.9.3082. PMID:10487669

Portaluppi F, Cortelli P, Avoni P et al. (1994). Diurnal blood pressure variation and hormonal correlates in fatal familial insomnia. Hypertension, 23:569–576. PMID:8175163

Pozo D, García-Mauriño S, Guerrero JM, Calvo JR (2004). mRNA expression of nuclear receptor RZR/RORalpha, melatonin membrane receptor MT, and hydroxindole-O-methyltransferase in different populations of human immune cells. J Pineal Res, 37:48–54 doi:10.1111/j.1600-079X.2004.00135.x. PMID:15230868

Preitner N, Damiola F, Lopez-Molina L et al. (2002). The orphan nuclear receptor REV-ERBalpha controls circadian transcription within the positive limb of the mammalian circadian oscillator. Cell, 110:251–260 doi:10.1016/S0092-8674(02)00825-5. PMID:12150932

Prendergast BJ, Hotchkiss AK, Nelson RJ (2003). Photoperiodic regulation of circulating leukocytes in juvenile Siberian hamsters: mediation by melatonin and testosterone. J Biol Rhythms, 18:473–480 doi:10.1177/0748730403258486. PMID:14667148

Pruessner JC, Wolf OT, Hellhammer DH et al. (1997). Free cortisol levels after awakening: a reliable biological marker for the assessment of adrenocortical activity. Life Sci, 61:2539–2549 doi:10.1016/S0024-3205(97)01008-4. PMID:9416776

Puig-Domingo M, Webb SM, Serrano J et al. (1992). Brief report: melatonin-related hypogonadotropic hypogonadism. N Engl J Med, 327:1356–1359 doi:10.1056/NEJM199211053271905. PMID:1406837

Putilov AA, Danilenko KV, Protopopova AY, Kripke DF (2002). Menstrual phase response to nocturnal light. Biol Rhythm Res, 33:23–38 doi:10.1076/brhm.33.1.23.1324.

Quera-Salva MA, Defrance R, Claustrat B et al. (1996). Rapid shift in sleep time and acrophase of melatonin secretion in short shift work schedule. Sleep, 19:539–543. PMID:8899932

Quera-Salva MA, Guilleminault C, Claustrat B et al. (1997). Rapid shift in peak melatonin secretion associated with improved performance in short shift work schedule. Sleep, 20:1145–1150. PMID:9493924

Raghavendra V, Singh V, Kulkarni SK, Agrewala JN (2001b). Melatonin enhances Th2 cell mediated immune responses: lack of sensitivity to reversal by naltrexone or benzodiazepine receptor antagonists. Mol Cell Biochem, 221:57–62 doi:10.1023/A:1010968611716. PMID:11506187

Raghavendra V, Singh V, Shaji AV et al. (2001a). Melatonin provides signal 3 to unprimed CD4(+) T cells but failed to stimulate LPS primed B cells. Clin Exp Immunol, 124:414–422 doi:10.1046/j.1365-2249.2001.01519.x. PMID:11472402

Rai S, Haldar C (2003). Pineal control of immune status and hematological changes in blood and bone marrow of male squirrels (Funambulus pennanti) during their reproductively active phase.

SHIFTWORK 747

Comp Biochem Physiol C Toxicol Pharmacol, 136:319–328 doi:10.1016/j.cca.2003.10.008. PMID:15012903

Ralph MR, Foster RG, Davis FC, Menaker M (1990). Transplanted suprachiasmatic nucleus determines circadian period. Science, 247:975–978 doi:10.1126/science.2305266. PMID:2305266

Ram PT, Dai J, Yuan L et al. (2002). Involvement of the mt1 melatonin receptor in human breast cancer. Cancer Lett, 179:141–150 doi:10.1016/S0304-3835(01)00873-4. PMID:11888668

Ramírez F, Fowell DJ, Puklavec M et al. (1996). Glucocorticoids promote a TH2 cytokine response by CD4+ T cells in vitro. J Immunol, 156:2406–2412. PMID:8786298

Rato AG, Pedrero JG, Martinez MA et al. (1999). Melatonin blocks the activation of estrogen receptor for DNA binding. FASEB J, 13:857–868. PMID:10224229

Reaven GM, Lithell H, Landsberg L (1996). Hypertension and associated metabolic abnormalities–the role of insulin resistance and the sympathoadrenal system. N Engl J Med, 334:374–381 doi:10.1056/NEJM199602083340607. PMID:8538710

Rechtschaffen A, Bergmann BM (1995). Sleep deprivation in the rat by the disk-over-water method. Behav Brain Res, 69:55–63 doi:10.1016/0166-4328(95)00020-T. PMID:7546318

Reddy AB, Field MD, Maywood ES, Hastings MH (2002). Differential resynchronisation of circadian clock gene expression within the suprachiasmatic nuclei of mice subjected to experimental jet lag. J Neurosci, 22:7326–7330. PMID:12196553

Reddy AB, Wong GKY, O’Neill J et al. (2005). Circadian clocks: neural and peripheral pacemakers that impact upon the cell division cycle. Mutat Res, 574:76–91. PMID:15914209

Reddy P, Zehring WA, Wheeler DA et al. (1984). Molecular analysis of the period locus in Drosophila melanogaster and identification of a transcript involved in biological rhythms. Cell, 38:701–710 doi:10.1016/0092-8674(84)90265-4. PMID:6435882

Reick M, Garcia JA, Dudley C, McKnight SL (2001). NPAS2: an analog of clock operative in the mammalian forebrain. Science, 293:506–509 doi:10.1126/science.1060699. PMID:11441147

Reilly JJ, Armstrong J, Dorosty AR et al.Avon Longitudinal Study of Parents and Children Study Team (2005). Early life risk factors for obesity in childhood: cohort study. BMJ, 330:1357–1364 doi:10.1136/bmj.38470.670903.E0. PMID:15908441

Reiter RJ (1972). Evidence for refractoriness of the pituitary-gonadal axis to the pineal gland in golden hamsters and its possible implications in annual reproductive rhythms. Anat Rec, 173:365–371 doi:10.1002/ar.1091730311. PMID:5039088

Reiter RJ (1980). The pineal and its hormones in the control of reproduction in mammals. Endocr Rev, 1:109–131 doi:10.1210/edrv-1-2-109. PMID:6263600

Reiter RJ (1991). Pineal melatonin: cell biology of its synthesis and of its physiological interactions. Endocr Rev, 12:151–180 doi:10.1210/edrv-12-2-151. PMID:1649044

Reiter RJ (2001) Reactive oxygen species, DNA damage and carcinogenesis. In: Bartsch C, Bartsch H, Blask DE, et al., Eds. Pineal Gland and Cancer. Neuroimmunoendocrine Mechanisms in Malignancy. Heidelberg: Springer. pp. 442–455.

Reiter RJ, Acuña-Castroviejo D, Tan DX, Burkhardt S (2001). Free radical-mediated molecular damage. Mechanisms for the protective actions of melatonin in the central nervous system. Ann N Y Acad Sci, 939:200–215 doi:10.1111/j.1749-6632.2001.tb03627.x. PMID:11462772

Reppert SM, Godson C, Mahle CD et al. (1995). Molecular characterization of a second melatonin receptor expressed in human retina and brain: the Mel1b melatonin receptor. Proc Natl Acad Sci USA, 92:8734–8738 doi:10.1073/pnas.92.19.8734. PMID:7568007


Reppert SM, Weaver DR (2001). Molecular analysis of mammalian circadian rhythms. Annu Rev Physiol, 63:647–676 doi:10.1146/annurev.physiol.63.1.647. PMID:11181971

Reppert SM, Weaver DR (2002). Coordination of circadian timing in mammals. Nature, 418:935–941 doi:10.1038/nature00965. PMID:12198538

Reppert SM, Weaver DR, Ebisawa T (1994). Cloning and characterization of a mammalian melatonin receptor that mediates reproductive and circadian responses. Neuron, 13:1177–1185 doi:10.1016/0896-6273(94)90055-8. PMID:7946354

Revell VL, Arendt J, Fogg LF, Skene DJ (2006). Alerting effects of light are sensitive to very short wavelengths. Neurosci Lett, 399:96–100 doi:10.1016/j.neulet.2006.01.032. PMID:16490309

Revell VL, Eastman CI (2005). How to trick mother nature into letting you fly around or stay up all night. J Biol Rhythms, 20:353–365 doi:10.1177/0748730405277233. PMID:16077154

Ribak J, Ashkenazi IE, Klepfish A et al. (1983). Diurnal rhythmicity and Air Force flight accidents due to pilot error. Aviat Space Environ Med, 54:1096–1099. PMID:6686440

Richard D, Huang Q, Timofeeva E (2000). The corticotropin-releasing hormone system in the regulation of energy balance in obesity. Int J Obes Relat Metab Disord, 24 Suppl 2;S36–S39. PMID:10997606

Rivest RW, Schulz P, Lustenberger S, Sizonenko PC (1989). Differences between circadian and ultradian organization of cortisol and melatonin rhythms during activity and rest. J Clin Endocrinol Metab, 68:721–729 doi:10.1210/jcem-68-4-721. PMID:2921307

Rivier C (1993). Neuroendocrine effects of cytokines in the rat. Rev Neurosci, 4:223–237. PMID:9155864

Roach GD, Burgess H, Lamond N et al. (2001). A week of simulated night work delays salivary melatonin onset. J Hum Ergol (Tokyo), 30:255–260. PMID:14564892

Roach GD, Lamond N, Dorrian J et al. (2005). Changes in the concentration of urinary 6-sulphatoxymelatonin during a week of simulated night work. Ind Health, 43:193–196 doi:10.2486/indhealth.43.193. PMID:15732322

Robilliard DL, Archer SN, Arendt J et al. (2002). The 3111 Clock gene polymorphism is not associated with sleep and circadian rhythmicity in phenotypically characterized human subjects. J Sleep Res, 11:305–312 doi:10.1046/j.1365-2869.2002.00320.x. PMID:12464098

Robinson JE, Kaynard AH, Karsch FJ (1986). Does melatonin alter pituitary responsiveness to gonadotropin-releasing hormone in the ewe? Neuroendocrinology, 43:635–640 doi:10.1159/000124593. PMID:3531906

Roky R, Obál F Jr, Valatx JL et al. (1995). Prolactin and rapid eye movement sleep regulation. Sleep, 18:536–542. PMID:8552923

Romagnani S (1996). Development of Th 1- or Th 2-dominated immune responses: what about the polarizing signals? Int J Clin Lab Res, 26:83–98 doi:10.1007/BF02592350. PMID:8856361

Ronkainen H, Pakarinen A, Kirkinen P, Kauppila A (1985). Physical exercise-induced changes and season-associated differences in the pituitary-ovarian function of runners and joggers. J Clin Endocrinol Metab, 60:416–422 doi:10.1210/jcem-60-3-416. PMID:3919040

Rönnberg L, Kauppila A, Leppaluoto J et al. (1990). Circadian and seasonal variation in human preovulatory follicular fluid melatonin concentration. J Clin Endocrinol Metab, 71:492–496 doi:10.1210/jcem-71-2-493. PMID: 2380343

Rosołowska-Huszcz D, Thaela MJ, Jagura M et al. (1991). Pineal influence on the diurnal rhythm of nonspecific immunity indices in chickens. J Pineal Res, 10:190–195 doi:10.1111/j.1600-079X.1991.tb00815.x. PMID:1920042

SHIFTWORK 749

Ross JK, Arendt J, Horne J, Haston W (1995). Night-shift work in Antarctica: sleep characteristics and bright light treatment. Physiol Behav, 57:1169–1174 doi:10.1016/0031-9384(95)00018-E. PMID:7652039

Ruby NF, Brennan TJ, Xie X et al. (2002). Role of melanopsin in circadian responses to light. Science, 298:2211–2213 doi:10.1126/science.1076701. PMID:12481140

Sack RL, Blood ML, Lewy AJ (1992). Melatonin rhythms in night shift workers. Sleep, 15:434–441. PMID:1455127

Sack RL, Brandes RW, Kendall AR, Lewy AJ (2000). Entrainment of free-running circadian rhythms by melatonin in blind people. N Engl J Med, 343:1070–1077 doi:10.1056/NEJM200010123431503. PMID:11027741

Sack RL, Hughes RJ, Edgar DM, Lewy AJ (1997). Sleep-promoting effects of melatonin: at what dose, in whom, under what conditions, and by what mechanisms? Sleep, 20:908–915. PMID:9415954

Sahar S, Sassone-Corsi P (2007). Circadian clock and breast cancer: a molecular link. Cell Cycle, 6:1329–1331. PMID:17534151

Sakami S, Ishikawa T, Kawakami N et al. (2002–2003). Coemergence of insomnia and a shift in the Th1/Th2 balance toward Th2 dominance. Neuroimmunomodulation, 10:337–343 doi:10.1159/000071474. PMID:12907840

Sakamoto K, Ishida N (2000). Light-induced phase-shifts in the circadian expression rhythm of mammalian period genes in the mouse heart. Eur J Neurosci, 12:4003–4006 doi:10.1046/j.1460-9568.2000.00302.x. PMID:11069596

Sakamoto K, Nagase T, Fukui H et al. (1998). Multitissue circadian expression of rat period homolog (rPer2) mRNA is governed by the mammalian circadian clock, the suprachiasmatic nucleus in the brain. J Biol Chem, 273:27039–27042 doi:10.1074/jbc.273.42.27039. PMID:9765215

Samuels MH, Lillehei K, Kleinschmidt-Demasters BK et al. (1990). Patterns of pulsatile pituitary glycoprotein secretion in central hypothyroidism and hypogonadism. J Clin Endocrinol Metab, 70:391–395 doi:10.1210/jcem-70-2-391. PMID:2105331

San Martin M, Touitou Y (2000). DHEA-sulfate causes a phase-dependent increase in melatonin secretion: a study of perifused rat pineal glands. Steroids, 65:491–496 doi:10.1016/S0039-128X(00)00111-2. PMID:10978727

Sánchez-Barceló EJ, Cos S, Fernández R, Mediavilla MD (2003). Melatonin and mammary cancer: a short review. Endocr Relat Cancer, 10:153–159 doi:10.1677/erc.0.0100153. PMID:12790777

Sánchez de la Peña S (1993). The feedsideward of cephalo-adrenal immune interactions. Chronobiologia, 20:1–52. PMID:8354098

Sánchez de la Peña S, Halberg F, Halberg E et al. (1983a). Pineal modulation of ACTH 1–17 effect upon murine corticosterone production. Brain Res Bull, 11:117–125 doi:10.1016/0361-9230(83)90060-6. PMID:6313140

Sánchez de la Peña S, Halberg F, Ungar F et al. (1983b). Circadian pineal modulation of pituitary effect on murine corticosterone in vitro. Brain Res Bull, 10:559–565 doi:10.1016/0361-9230(83)90155-7. PMID:6860981

Sandahl B (1978). Seasonal birth pattern in Sweden in relation to birth order and maternal age. Acta Obstet Gynecol Scand, 57:393–396 doi:10.3109/00016347809156517. PMID:726872

Sassin JF, Frantz AG, Kapen S, Weitzman ED (1973). The nocturnal rise of human prolactin is dependent on sleep. J Clin Endocrinol Metab, 37:436–440 doi:10.1210/jcem-37-3-436. PMID:4361974


Sassin JF, Frantz AG, Weitzman ED, Kapen S (1972). Human prolactin: 24-hour pattern with increased release during sleep. Science, 177:1205–1207 doi:10.1126/science.177.4055.1205. PMID:5057627

Sawchenko PE, Swanson LW (1985). Localization, colocalization, and plasticity of corticotropin-releasing factor immunoreactivity in rat brain. Fed Proc, 44:221–227. PMID:2981743

Schernhammer ES, Rosner B, Willett WC et al. (2004). Epidemiology of urinary melatonin in women and its relation to other hormones and night work. Cancer Epidemiol Biomarkers Prev, 13:936–943. PMID:15184249

Schibler U, Ripperger J, Brown SA (2003). Peripheral circadian oscillators in mammals: time and food. J Biol Rhythms, 18:250–260 doi:10.1177/0748730403018003007. PMID:12828282

Sekaran S, Foster RG, Lucas RJ, Hankins MW (2003). Calcium imaging reveals a network of intrinsically light-sensitive inner-retinal neurons. Curr Biol, 13:1290–1298 doi:10.1016/S0960-9822(03)00510-4. PMID:12906788

Sekine M, Yamagami T, Handa K et al. (2002). A dose-response relationship between short sleeping hours and childhood obesity: results of the Toyama Birth Cohort Study. Child Care Health Dev, 28:163–170 doi:10.1046/j.1365-2214.2002.00260.x. PMID:11952652

Serda SM, Wei ET (1992). Epinephrine-induced pulmonary oedema in rats is inhibited by corticotropin-releasing factor. Pharmacol Res, 26:85–91 doi:10.1016/1043-6618(92)90708-J. PMID:1513751

Sharma M, Palacios-Bois J, Schwartz G et al. (1989). Circadian rhythms of melatonin and cortisol in aging. Biol Psychiatry, 25:305–319 doi:10.1016/0006-3223(89)90178-9. PMID:2914154

Shearman LP, Zylka MJ, Weaver DR et al. (1997). Two period homologs: circadian expression and photic regulation in the suprachiasmatic nuclei. Neuron, 19:1261–1269 doi:10.1016/S0896-6273(00)80417-1. PMID:9427249

Shih M-C, Yeh K-T, Tang K-P et al. (2006). Promoter methylation in circadian genes of endometrial cancers detected by methylation-specific PCR. Mol Carcinog, 45:732–740 doi:10.1002/mc.20198. PMID:16683245

Simard M, Pekary AE, Smith VP, Hershman JM (1989). Thyroid hormones modulate thyrotropin-releasing hormone biosynthesis in tissues outside the hypothalamic-pituitary axis of male rats. Endocrinology, 125:524–531 doi:10.1210/endo-125-1-524. PMID:2500333

Simpson HW, Candlish W, Pauson AW et al. (1988). Genesis of breast cancer is in the premenopause. Lancet, 332:74–76 doi:10.1016/S0140-6736(88)90007-4. PMID:2898701

Simpson HW, Pauson A, Cornélissen G (1989). The chronopathology of breast pre-cancer. Chronobiologia, 16:365–372. PMID:2627819

Skene DJ, Lockley SW, Thapan K, Arendt J (1999). Effects of light on human circadian rhythms. Reprod Nutr Dev, 39:295–304 doi:10.1051/rnd:19990302. PMID:10420432

Smith EM, Morrill AC, Meyer WJ 3rd, Blalock JE (1986). Corticotropin releasing factor induction of leukocyte-derived immunoreactive ACTH and endorphins. Nature, 321:881–882 doi:10.1038/321881a0. PMID:3014342

Smith KA, Schoen MW, Czeisler CA (2004). Adaptation of human pineal melatonin suppression by recent photic history. J Clin Endocrinol Metab, 89:3610–3614 doi:10.1210/jc.2003-032100. PMID:15240654

Soszyński P, Stowińska-Srzednicka J, Kasperlik-Zatuska A, Zgliczyński S (1989). Decreased melatonin concentration in Cushing’s syndrome. Horm Metab Res, 21:673–674 doi:10.1055/s-2007-1009317. PMID:2559013

SHIFTWORK 751

Späth-Schwalbe E, Gofferje M, Kern W et al. (1991). Sleep disruption alters nocturnal ACTH and cortisol secretory patterns. Biol Psychiatry, 29:575–584 doi:10.1016/0006-3223(91)90093-2. PMID:1647222

Spencer CA, Greenstadt MA, Wheeler WS et al. (1980). The influence of long-term low dose thyrotropin-releasing hormone infusions on serum thyrotropin and prolactin concentrations in man. J Clin Endocrinol Metab, 51:771–775 doi:10.1210/jcem-51-4-771. PMID:6774994

Spiegel K, Follenius M, Simon C et al. (1994). Prolactin secretion and sleep. Sleep, 17:20–27. PMID:8191199

Spiegel K, Leproult R, Colecchia EF et al. (2000). Adaptation of the 24-h growth hormone profile to a state of sleep debt. Am J Physiol Regul Integr Comp Physiol, 279:R874–R883. PMID:10956244

Spiegel K, Leproult R, L’hermite-Balériaux M et al. (2004a). Leptin levels are dependent on sleep duration: relationships with sympathovagal balance, carbohydrate regulation, cortisol, and thyrotropin. J Clin Endocrinol Metab, 89:5762–5771 doi:10.1210/jc.2004-1003. PMID:15531540

Spiegel K, Leproult R, Van Cauter E (1999). Impact of sleep debt on metabolic and endocrine function. Lancet, 354:1435–1439 doi:10.1016/S0140-6736(99)01376-8. PMID:10543671

Spiegel K, Leproult R, Van Cauter E (2003) [Impact of sleep debt on physiological rhythms.] Rev Neurol (Paris), 159 (Suppl.11):6S11–20. PMID: 14646794

Spiegel K, Luthringer R, Follenius M et al. (1995). Temporal relationship between prolactin secretion and slow-wave electroencephalic activity during sleep. Sleep, 18:543–548. PMID:8552924

Spiegel K, Tasali E, Penev P, Van Cauter E (2004b). Brief communication: Sleep curtailment in healthy young men is associated with decreased leptin levels, elevated ghrelin levels, and increased hunger and appetite. Ann Intern Med, 141:846–850. PMID:15583226

Stehle J, Vanecek J, Vollrath L (1989). Effects of melatonin on spontaneous electrical activity of neurons in rat suprachiasmatic nuclei: an in vitro iontophoretic study. J Neural Transm, 78:173–177 doi:10.1007/BF01252503. PMID:2809584

Stetson MH, Matt KS, Watson-Whitmyre M (1976). Photoperiodism and reproduction in golden hamsters: circadian organization and the termination of photorefractoriness. Biol Reprod, 14:531–537 doi:10.1095/biolreprod14.5.531. PMID:1276317

Stevens RG (1987). Electric power use and breast cancer: a hypothesis. Am J Epidemiol, 125:556–561. PMID:3548332

Stevens RG (2006). Artificial lighting in the industrialized world: circadian disruption and breast cancer. Cancer Causes Control, 17:501–507 doi:10.1007/s10552-005-9001-x. PMID:16596303

Stokkan K-A, Yamazaki S, Tei H et al. (2001). Entrainment of the circadian clock in the liver by feeding. Science, 291:490–493 doi:10.1126/science.291.5503.490. PMID:11161204

Storch K-F, Lipan O, Leykin I et al. (2002). Extensive and divergent circadian gene expression in liver and heart. Nature, 417:78–83 doi:10.1038/nature744. PMID:11967526

Stossi F, Likhite VS, Katzenellenbogen JA, Katzenellenbogen BS (2006). Estrogen-occupied estrogen receptor represses cyclin G2 gene expression and recruits a repressor complex at the cyclin G2 promoter. J Biol Chem, 281:16272–16278 doi:10.1074/jbc.M513405200. PMID:16608856

Suh BY, Liu JH, Berga SL et al. (1988). Hypercortisolism in patients with functional hypothalamic-amenorrhea. J Clin Endocrinol Metab, 66:733–739 doi:10.1210/jcem-66-4-733. PMID:3346352


Taheri S (2006). The link between short sleep duration and obesity: we should recommend more sleep to prevent obesity. Arch Dis Child, 91:881–884 doi:10.1136/adc.2005.093013. PMID:17056861

Taheri S, Lin L, Austin D et al. (2004). Short sleep duration is associated with reduced leptin, elevated ghrelin, and increased body mass index. PLoS Med, 1:e62 doi:10.1371/journal.pmed.0010062. PMID: 15602591

Takano A, Shimizu K, Kani S et al. (2000). Cloning and characterization of rat casein kinase 1epsilon. FEBS Lett, 477:106–112 doi:10.1016/S0014-5793(00)01755-5. PMID:10899319

Tamarkin L, Baird CJ, Almeida OFX (1985). Melatonin: a coordinating signal for mammalian reproduction? Science, 227:714–720 doi:10.1126/science.3881822. PMID:3881822

Tannenbaum MG, Reiter RJ, Vaughan MK et al. (1987). Adrenalectomy prevents changes in rat pineal melatonin content and N-acetyltransferase activity induced by acute insulin stress. J Pineal Res, 4:395–402 doi:10.1111/j.1600-079X.1987.tb00879.x. PMID:3312570

Tei H, Okamura H, Shigeyoshi Y et al. (1997). Circadian oscillation of a mammalian homologue of the Drosophila period gene. Nature, 389:512–516 doi:10.1038/39086. PMID:9333243

Terzolo M, Piovesan A, Puligheddu B et al. (1990). Effects of long-term, low-dose, time-specified melatonin administration on endocrine and cardiovascular variables in adult men. J Pineal Res, 9:113–124 doi:10.1111/j.1600-079X.1990.tb00699.x. PMID:2177501

Terzolo M, Revelli A, Guidetti D et al. (1993). Evening administration of melatonin enhances the pulsatile secretion of prolactin but not of LH and TSH in normally cycling women. Clin Endocrinol (Oxf), 39:185–191 doi:10.1111/j.1365-2265.1993.tb01772.x. PMID:8370131

Tetsuo M, Poth M, Markey SP (1982). Melatonin metabolite excretion during childhood and puberty. J Clin Endocrinol Metab, 55:311–313 doi:10.1210/jcem-55-2-311. PMID:7085857

Thapan K, Arendt J, Skene DJ (2001). An action spectrum for melatonin suppression: evidence for a novel non-rod, non-cone photoreceptor system in humans. J Physiol, 535:261–267 doi:10.1111/j.1469-7793.2001.t01-1-00261.x. PMID:11507175

Ticher A, Haus E, Ron IG et al. (1996). The pattern of hormonal circadian time structure (acrophase) as an assessor of breast-cancer risk. Int J Cancer, 65:591–593 doi:10.1002/(SICI)1097-0215(19960301)65:5<591::AID-IJC6>3.0.CO;2-Y. PMID:8598308

Timonen S, Franzas B, Wichmann K (1964). Photosensibility of the human pituitary. Ann Chir Gynaecol Fenn, 53:165–172. PMID:14163867

Torres-Farfan C, Richter HG, Rojas-García P et al. (2003). mt1 Melatonin receptor in the primate adrenal gland: inhibition of adrenocorticotropin-stimulated cortisol production by melatonin. J Clin Endocrinol Metab, 88:450–458 doi:10.1210/jc.2002-021048. PMID:12519889

Torres-Farfan C, Rocco V, Monsó C et al. (2006). Maternal melatonin effects on clock gene expression in a nonhuman primate fetus. Endocrinology, 147:4618–4626 doi:10.1210/en.2006-0628. PMID:16840546

Tortosa F, Puig-Domingo M, Peinado MA et al. (1989). Enhanced circadian rhythm of melatonin in anorexia nervosa. Acta Endocrinol (Copenh), 120:574–578. PMID:2728803

Touitou Y, Fèvre M, Bogdan A et al. (1984). Patterns of plasma melatonin with ageing and mental condition: stability of nyctohemeral rhythms and differences in seasonal variations. Acta Endocrinol (Copenh), 106:145–151. PMID:6539550

Touitou Y, Haus E (1992) Biological rhythms and aging. In: Touitou Y & Haus E, Eds. Biologic Rhythms in Clinical and Laboratory Medicine. Berling, Heidelberg: Springer-Verlag. pp. 188–207.

SHIFTWORK 753

Touitou Y, Motohashi Y, Reinberg A et al. (1990). Effect of shift work on the night-time secretory patterns of melatonin, prolactin, cortisol and testosterone. Eur J Appl Physiol Occup Physiol, 60:288–292 doi:10.1007/BF00379398. PMID:2357985

Tricoire H, Møller M, Chemineau P, Malpaux B (2003). Origin of cerebrospinal fluid melatonin and possible function in the integration of photoperiod. Reprod Suppl, 61:311–321. PMID:14635944

Ungar F, Halberg F (1962). Circadian rhythm in the in vitro response of mouse adrenal to adrenocorticotropic hormone. Science, 137:1058–1060 doi:10.1126/science.137.3535.1058. PMID:13923797

van Amelsvoort LG, Schouten EG, Kok FJ (1999). Duration of shiftwork related to body mass index and waist to hip ratio. Int J Obes Relat Metab Disord, 23:973–978 doi:10.1038/sj.ijo.0801028. PMID:10490804

Van Cauter E, Copinschi G (2000). Interrelationships between growth hormone and sleep. Growth Horm IGF Res, 10 Suppl B;S57–S62 doi:10.1016/S1096-6374(00)80011-8. PMID:10984255

Van Cauter E, Kerkhofs M, Caufriez A et al. (1992). A quantitative estimation of growth hormone secretion in normal man: reproducibility and relation to sleep and time of day. J Clin Endocrinol Metab, 74:1441–1450 doi:10.1210/jc.74.6.1441. PMID:1592892

Van Cauter E, Plat L, Copinschi G (1998). Interrelations between sleep and the somatotropic axis. Sleep, 21:553–566. PMID:9779515

Van Cauter E, Plat L, Scharf MB et al. (1997). Simultaneous stimulation of slow-wave sleep and growth hormone secretion by gamma-hydroxybutyrate in normal young Men. J Clin Invest, 100:745–753 doi:10.1172/JCI119587. PMID:9239423

Van Cauter E, Sturis J, Byrne MM et al. (1993). Preliminary studies on the immediate phase-shifting effects of light and exercise on the human circadian clock. J Biol Rhythms, 8 Suppl;S99–S108. PMID:8274769

Van Cauter E, Sturis J, Byrne MM et al. (1994). Demonstration of rapid light-induced advances and delays of the human circadian clock using hormonal phase markers. Am J Physiol, 266:E953–E963. PMID:8023927

van Coevorden A, Mockel J, Laurent E et al. (1991). Neuroendocrine rhythms and sleep in aging men. Am J Physiol, 260:E651–E661. PMID:2018128

van de Wetering M, de Lau W, Clevers H (2002). WNT signaling and lymphocyte development. Cell, 109 Suppl;S13–S19 doi:10.1016/S0092-8674(02)00709-2. PMID:11983149

Vanĕcek J (1988). Melatonin binding sites. J Neurochem, 51:1436–1440 doi:10.1111/j.1471-4159.1988.tb01108.x. PMID:3171587

Vanecek J (1995). Melatonin inhibits increase of intracellular calcium and cyclic AMP in neonatal rat pituitary via independent pathways. Mol Cell Endocrinol, 107:149–153 doi:10.1016/0303-7207(94)03437-X. PMID:7768326

Vanecek J, Klein DC (1995). Melatonin inhibition of GnRH-induced LH release from neonatal rat gonadotroph: involvement of Ca2+ not cAMP. Am J Physiol, 269:E85–E90. PMID:7631782

Vaughan GM, Reiter RJ (1986). Pineal dependence of the Syrian hamster’s nocturnal serum melatonin surge. J Pineal Res, 3:9–14 doi:10.1111/j.1600-079X.1986.tb00721.x. PMID:3958897

Vaughan MK, Hubbard GB, Champney TH et al. (1987). Splenic hypertrophy and extramedullary hematopoiesis induced in male Syrian hamsters by short photoperiod or melatonin injections and reversed by melatonin pellets or pinealectomy. Am J Anat, 179:131–136 doi:10.1002/aja.1001790205. PMID:3618524


Vaughan MK, Reiter RJ (1971). Transient hypertrophy of the ventral prostate and coagulating glands and accelerated thymic involution following pinealectomy in the mouse. Tex Rep Biol Med, 29:579–586. PMID:4113015

Veldhuis JD, Evans WS, Demers LM et al. (1985). Altered neuroendocrine regulation of gonadotropin secretion in women distance runners. J Clin Endocrinol Metab, 61:557–563 doi:10.1210/jcem-61-3-557. PMID:3926810

Veldhuis JD, Johnson ML, Lizarralde G, Iranmanesh A (1992). Rhythmic and nonrhythmic modes of anterior pituitary gland secretion. Chronobiol Int, 9:371–379 doi:10.3109/07420529209064549. PMID:1394609

Verkasalo PK, Pukkala E, Stevens RG et al. (1999). Inverse association between breast cancer incidence and degree of visual impairment in Finland. Br J Cancer, 80:1459–1460 doi:10.1038/sj.bjc.6690544. PMID:10424751

Vermeulen M, Palermo M, Giordano M (1993). Neonatal pinealectomy impairs murine antibody-dependent cellular cytotoxicity. J Neuroimmunol, 43:97–101 doi:10.1016/0165-5728(93)90079-E. PMID:8458988

Vgontzas AN, Bixler EO, Lin HM et al. (2001). Chronic insomnia is associated with nyctohemeral activation of the hypothalamic-pituitary-adrenal axis: clinical implications. J Clin Endocrinol Metab, 86:3787–3794 doi:10.1210/jc.86.8.3787. PMID:11502812

Vgontzas AN, Chrousos GP (2002). Sleep, the hypothalamic-pituitary-adrenal axis, and cytokines: multiple interactions and disturbances in sleep disorders. Endocrinol Metab Clin North Am, 31:15–36 doi:10.1016/S0889-8529(01)00005-6. PMID:12055986

Vgontzas AN, Zoumakis E, Bixler EO et al. (2004). Adverse effects of modest sleep restriction on sleepiness, performance, and inflammatory cytokines. J Clin Endocrinol Metab, 89:2119–2126 doi:10.1210/jc.2003-031562. PMID:15126529

Vidaller A, Guadarrama F, Llorente L et al. (1992). Hyperprolactinemia inhibits natural killer (NK) cell function in vivo and its bromocriptine treatment not only corrects it but makes it more efficient. J Clin Immunol, 12:210–215 doi:10.1007/BF00918091. PMID:1400902

Vidaller A, Llorente L, Larrea F et al. (1986). T-cell dysregulation in patients with hyperprolactinemia: effect of bromocriptine treatment. Clin Immunol Immunopathol, 38:337–343 doi:10.1016/0090-1229(86)90243-6. PMID:2935343

Viola AU, Archer SN, James LM et al. (2007). PER3 polymorphism predicts sleep structure and waking performance. Curr Biol, 17:613–618 doi:10.1016/j.cub.2007.01.073. PMID:17346965

Vioque J, Torres A, Quiles J (2000). Time spent watching television, sleep duration and obesity in adults living in Valencia, Spain. Int J Obes Relat Metab Disord, 24:1683–1688 doi:10.1038/sj.ijo.0801434. PMID:11126224

Visser J, van Boxel-Dezaire A, Methorst D et al. (1998). Differential regulation of interleukin-10 (IL-10) and IL-12 by glucocorticoids in vitro. Blood, 91:4255–4264. PMID:9596674

Vitaterna MH, King DP, Chang AM et al. (1994). Mutagenesis and mapping of a mouse gene, Clock, essential for circadian behavior. Science, 264:719–725 doi:10.1126/science.8171325. PMID:8171325

von Gall C, Weaver DR, Moek J et al. (2005). Melatonin plays a crucial role in the regulation of rhythmic clock gene expression in the mouse pars tuberalis. Ann N Y Acad Sci, 1040:508–511 doi:10.1196/annals.1327.105. PMID:15891103

SHIFTWORK 755

Vondrasová D, Hájek I, Illnerová H (1997). Exposure to long summer days affects the human melatonin and cortisol rhythms. Brain Res, 759:166–170 doi:10.1016/S0006-8993(97)00358-2. PMID:9219878

Voordouw BC, Euser R, Verdonk RE et al. (1992). Melatonin and melatonin-progestin combinations alter pituitary-ovarian function in women and can inhibit ovulation. J Clin Endocrinol Metab, 74:108–117 doi:10.1210/jc.74.1.108. PMID:1727807

Waldhauser F, Boepple PA, Crowley WF Jr (1993) Changes of serum melatonin levels during human sexual maturation. In: Grave GD & Cutler GB, Jr, Eds. Sexual Precocity. New York: Raven Press. pp. 181–192.

Waldhauser F, Boepple PA, Schemper M et al. (1991). Serum melatonin in central precocious puberty is lower than in age-matched prepubertal children. J Clin Endocrinol Metab, 73:793–796 doi:10.1210/jcem-73-4-793. PMID:1909703

Waldhauser F, Dietzel M (1985). Daily and annual rhythms in human melatonin secretion: role in puberty control. Ann N Y Acad Sci, 453:205–214 doi:10.1111/j.1749-6632.1985.tb11811.x. PMID:3865581

Waldhauser F, Steger H, Vorkapic P (1987). Melatonin secretion in man and the influence of exogenous melatonin on some physiological and behavioral variables. Adv Pineal Res, 2:207–223.

Waldhauser F, Weiszenbacher G, Frisch H et al. (1984). Fall in nocturnal serum melatonin during prepuberty and pubescence. Lancet, 323:362–365 doi:10.1016/S0140-6736(84)90412-4. PMID:6141425

Waldhauser F, Weiszenbacher G, Tatzer E et al. (1988). Alterations in nocturnal serum melatonin levels in humans with growth and aging. J Clin Endocrinol Metab, 66:648–652 doi:10.1210/jcem-66-3-648. PMID:3350912

Waldstreicher J, Duffy JF, Brown EN et al. (1996). Gender differences in the temporal organization of proclactin (PRL) secretion: evidence for a sleep-independent circadian rhythm of circulating PRL levels- a clinical research center study. J Clin Endocrinol Metab, 81:1483–1487 doi:10.1210/jc.81.4.1483. PMID:8636355

Walters JF, Hampton SM, Ferns GA, Skene DJ (2005). Effect of menopause on melatonin and alertness rhythms investigated in constant routine conditions. Chronobiol Int, 22:859–872 doi:10.1080/07420520500263193. PMID:16298772

Wang C, Fan S, Li Z et al. (2005). Cyclin D1 antagonizes BRCA1 repression of estrogen receptor alpha activity. Cancer Res, 65:6557–6567 doi:10.1158/0008-5472.CAN-05-0486. PMID:16061635

Warman VL, Dijk D-J, Warman GR et al. (2003). Phase advancing human circadian rhythms with short wavelength light. Neurosci Lett, 342:37–40 doi:10.1016/S0304-3940(03)00223-4. PMID:12727312

Weaver DR, Rivkees SA, Reppert SM (1989). Localization and characterization of melatonin receptors in rodent brain by in vitro autoradiography. J Neurosci, 9:2581–2590. PMID:2545841

Webley GE, Leidenberger F (1986). The circadian pattern of melatonin and its positive relationship with progesterone in women. J Clin Endocrinol Metab, 63:323–328 doi:10.1210/jcem-63-2-323. PMID:3722324

Webley GE, Luck MR (1986). Melatonin directly stimulates the secretion of progesterone by human and bovine granulosa cells in vitro. J Reprod Fertil, 78:711–717 doi:10.1530/jrf.0.0780711. PMID:3806524


Wehr TA (1991). The durations of human melatonin secretion and sleep respond to changes in daylength (photoperiod). J Clin Endocrinol Metab, 73:1276–1280 doi:10.1210/jcem-73-6-1276. PMID:1955509

Wehr TA (1996). A ‘clock for all seasons’ in the human brain. Prog Brain Res, 111:321–342 doi:10.1016/S0079-6123(08)60416-1. PMID:8990923

Wehr TA (2001). Photoperiodism in humans and other primates: evidence and implications. J Biol Rhythms, 16:348–364 doi:10.1177/074873001129002060. PMID:11506380

Wehr TA, Giesen HA, Moul DE et al. (1995). Suppression of men’s responses to seasonal changes in day length by modern artificial lighting. Am J Physiol, 269:R173–R178. PMID:7631890

Wei ET, Gao GC (1991). Corticotropin-releasing factor: an inhibitor of vascular leakage in rat skeletal muscle and brain cortex after injury. Regul Pept, 33:93–104 doi:10.1016/0167-0115(91)90205-U. PMID:1882088

Weibel L, Follenius M, Spiegel K et al. (1995). Comparative effect of night and daytime sleep on the 24-hour cortisol secretory profile. Sleep, 18:549–556. PMID:8552925

Weibel L, Follenius M, Spiegel K et al. (1997). Growth hormone secretion in night workers. Chronobiol Int, 14:49–60 doi:10.3109/07420529709040541. PMID:9042551

Weinberg U, Weitzman ED, Fukushima DK et al. (1980). Melatonin does not suppress the pituitary luteinizing hormone response to luteinizing hormone-releasing hormone in men. J Clin Endocrinol Metab, 51:161–162 doi:10.1210/jcem-51-1-161. PMID:6991518

Welsh DK, Logothetis DE, Meister M, Reppert SM (1995). Individual neurons dissociated from rat suprachiasmatic nucleus express independently phased circadian firing rhythms. Neuron, 14:697–706 doi:10.1016/0896-6273(95)90214-7. PMID:7718233

Werner S, Brismar K, Wetterberg L, Eneroth P (1981) Circadian rhythms of melatonin, prolactin, growth hormone and cortisol in patients with pituitary adenomas, empty sella syndrome, and Cushing’s syndrome due to adrenal tumours. In: Birau N & Schloot W, Eds. Melatonin: Current Status and Perspectives. Oxford: Pergamon Press Ltd. pp 357–364.

Wetterberg L, Arendt J, Paunier L et al. (1976). Human serum melatonin changes during the menstrual cycle. J Clin Endocrinol Metab, 42:185–188 doi:10.1210/jcem-42-1-185. PMID:1249188

Wilson DW, Griffiths K, Halberg F et al. (1983). Breast skin temperature rhythms in relation to ovulation. Chronobiologia, 10:231–243. PMID:6641367

Winget CM, Soliman MRI, Holley DC, Meylor JS (1992) Chronobiology of physical performance and sports medicine. In: Touitou Y & Haus E, Ed. Biologic Rhythms in Clinical and Laboratory Medicine. Berlin, Heidelberg, Paris: Springer. pp. 230–242.

Woo MM, Tai CJ, Kang SK et al. (2001). Direct action of melatonin in human granulosa-luteal cells. J Clin Endocrinol Metab, 86:4789–4797 doi:10.1210/jc.86.10.4789. PMID:11600542

Wright J, Aldhous M, Franey C et al. (1986). The effects of exogenous melatonin on endocrine function in man. Clin Endocrinol (Oxf), 24:375–382 doi:10.1111/j.1365-2265.1986.tb01641.x. PMID:3742833

Wright KP Jr, Czeisler CA (2002). Absence of circadian phase resetting in response to bright light behind the knees. Science, 297:571 doi:10.1126/science.1071697. PMID:12142528

Wright KP Jr, Gronfier C, Duffy JF, Czeisler CA (2005). Intrinsic period and light intensity determine the phase relationship between melatonin and sleep in humans. J Biol Rhythms, 20:168–177 doi:10.1177/0748730404274265. PMID:15834113

SHIFTWORK 757

Wright KP Jr, Hughes RJ, Kronauer RE et al. (2001). Intrinsic near-24-h pacemaker period determines limits of circadian entrainment to a weak synchronizer in humans. Proc Natl Acad Sci USA, 98:14027–14032 doi:10.1073/pnas.201530198. PMID:11717461

Wu WT, Chen YC, Reiter RJ (1988). Day-night differences in the response of the pineal gland to swimming stress. Proc Soc Exp Biol Med, 187:315–319. PMID:3347608

Wu Y-H, Zhou J-N, Balesar R et al. (2006). Distribution of MT1 melatonin receptor immunoreactivity in the human hypothalamus and pituitary gland: colocalization of MT1 with vasopressin, oxytocin, and corticotropin-releasing hormone. J Comp Neurol, 499:897–910 doi:10.1002/cne.21152. PMID:17072839

Wurtman RJ, Axelrod J, Potter LT (1964). The uptake of H3-melatonin in endocrine and nervous tissues and the effects of constant light exposure. J Pharmacol Exp Ther, 143:314–318. PMID:14161142

Yan L, Silver R (2002). Differential induction and localization of mPer1 and mPer2 during advancing and delaying phase shifts. Eur J Neurosci, 16:1531–1540 doi:10.1046/j.1460-9568.2002.02224.x. PMID:12405967

Yang M-Y, Chang J-G, Lin P-M et al. (2006). Downregulation of circadian clock genes in chronic myeloid leukemia: alternative methylation pattern of hPER3. Cancer Sci, 97:1298–1307 doi:10.1111/j.1349-7006.2006.00331.x. PMID:16999817

Yeh K-T, Yang M-Y, Liu T-C et al. (2005). Abnormal expression of period 1 (PER1) in endometrial carcinoma. J Pathol, 206:111–120 doi:10.1002/path.1756. PMID:15809976

Yie SM, Brown GM, Liu GY et al. (1995a). Melatonin and steroids in human pre-ovulatory follicular fluid: seasonal variations and granulosa cell steroid production. Hum Reprod, 10:50–55 doi:10.1093/humrep/10.1.50. PMID:7745070

Yie SM, Niles LP, Younglai EV (1995b). Melatonin receptors on human granulosa cell membranes. J Clin Endocrinol Metab, 80:1747–1749 doi:10.1210/jc.80.5.1747. PMID:7745030

Yoo S-H, Yamazaki S, Lowrey PL et al. (2004). PERIOD2:LUCIFERASE real-time reporting of circadian dynamics reveals persistent circadian oscillations in mouse peripheral tissues. Proc Natl Acad Sci USA, 101:5339–5346 doi:10.1073/pnas.0308709101. PMID:14963227

Yoon I-Y, Kripke DF, Elliott JA et al. (2003). Age-related changes of circadian rhythms and sleep-wake cycles. J Am Geriatr Soc, 51:1085–1091 doi:10.1046/j.1532-5415.2003.51356.x. PMID:12890070

Young MW, Kay SA (2001). Time zones: a comparative genetics of circadian clocks. Nat Rev Genet, 2:702–715 doi:10.1038/35088576. PMID:11533719

Youngstedt SD, Kripke DF, Elliott JA, Klauber MR (2001). Circadian abnormalities in older adults. J Pineal Res, 31:264–272 doi:10.1034/j.1600-079X.2001.310311.x. PMID:11589762

Zeitzer JM, Daniels JE, Duffy JF et al. (1999). Do plasma melatonin concentrations decline with age? Am J Med, 107:432–436 doi:10.1016/S0002-9343(99)00266-1. PMID:10569297

Zeitzer JM, Dijk DJ, Kronauer R et al. (2000). Sensitivity of the human circadian pacemaker to nocturnal light: melatonin phase resetting and suppression. J Physiol, 526:695–702. PMID:10922269

Zhao ZY, Touitou Y (1993). Kinetic changes of melatonin release in rat pineal perifusions at different circadian stages. Effects of corticosteroids. Acta Endocrinol (Copenh), 129:81–88. PMID:8351960

Zheng B, Larkin DW, Albrecht U et al. (1999). The mPer2 gene encodes a functional component of the mammalian circadian clock. Nature, 400:169–173 doi:10.1038/22659. PMID:10408444


Zhu Y, Brown HN, Zhang Y et al. (2005). Period3 structural variation: a circadian biomarker associated with breast cancer in young women. Cancer Epidemiol Biomarkers Prev, 14:268–270. PMID:15668506

Zhu Y, Leaderer D, Guss C et al. (2007). Ala394Thr polymorphism in the clock gene NPAS2: a circadian modifier for the risk of non-Hodgkin’s lymphoma. Int J Cancer, 120:432–435 doi:10.1002/ijc.22321. PMID:17096334

Zhu Y, Stevens RG, Leaderer D et al. (2008). Non-synonymous polymorphisms in the circadian gene NPAS2 and breast cancer risk. Breast Cancer Res Treat, 107:421–425 doi:10.1007/s10549-007-9565-0. PMID:17453337

Zlotos DP (2005). Recent advances in melatonin receptor ligands. Arch Pharm Chem Life Sci (Weinheim), 338:229–247 doi:10.1002/ardp.200400996. PMID:15952241

Zylka MJ, Shearman LP, Weaver DR, Reppert SM (1998). Three period homologs in mammals: differential light responses in the suprachiasmatic circadian clock and oscillating transcripts outside of brain. Neuron, 20:1103–1110 doi:10.1016/S0896-6273(00)80492-4. PMID:9655499

SHIFTWORK 759

5. Summary of Data Reported

5.1 Exposure data

Several types of shift systems exist, described according to several main characteristics: permanent or rotating, continuous (all days of the week are covered) or discontinuous (interruption at the weekend or on Sunday), or with or without night work. Other important organizational factors that may have an impact on health are: length of the shift cycle, duration of shifts, number of alternating workers/crews, start and finish time of the duty periods, speed and direction (clockwise or anticlockwise) of shift rotation, number and position of rest days and regularity/irregularity of the shift schedules.

The amount of night work in any shift period is the most important factor to be considered in the disruption of biological functions. The amount of sleep of the shiftworker decreases both in terms of quantity and quality, both on night shifts (due to circadian and environmental reasons) and on early-morning shifts.

Shiftwork, that includes night work, is estimated to involve about 15–20% of the total working population, although reliable and comparable statistics are not available in most countries. In Europe, large differences have been recorded among countries (from 6.4 to 30%), and between self-employed (5.7%) and employees (19.8%); in the USA, the average prevalence of shiftwork that includes night work is 14.8% (16.7% in men and 12.4% in women). Shiftwork is most prevalent among workers in the health care, transportation, communication, leisure and hospitality sectors (above 30%), and in the service, mining and industrial manufacturing sectors (20–30%). The prevalence of shiftwork is more common in work schedules of younger workers but decreases with the age of the workers, from more than 20% in the youngest decades of life to approximately 10% after 55 years of age.

At the time of writing, there is no known biomarker of exposure for shiftwork. However, because of the importance of melatonin in the relation to the activity in the circadian rhythm, levels of melatonin could be useful biomarkers of circadian disruption. Melatonin can be measured in the blood, saliva, or urine. The measurement of melatonin concentration in plasma at regular intervals (e.g. hourly) can identify the onset, offset and duration of melatonin secretion, the time at which peak secretion occurred and the total amount of melatonin secreted. Salivary melatonin concentration is a good alternative measure as it is highly correlated with serum concentrations. The primary urinary metabolite of melatonin, 6-sulfatoxymelatonin, may also be a useful biomarker.

The disruptive effects of night work on the biological functions and social life have been recognized by some regulators both at the national and international levels, i.e. the International Labour Organization, the European Union, and the Federal Aviation Administration.


5.2 Human carcinogenicity data

Female breast cancer

Eight studies from various geographic regions have been designed to assess the relationship between breast cancer and shiftwork that involves night work. Six of these eight studies, including two prospective cohort studies in nurses, have consistently pointed towards a modestly increased risk of breast cancer among long-term employees who performed night shiftwork, defined in different ways. Most studies reported this increased risk after controlling for potential confounders. Two of the eight studies, one of which appeared to be hampered by important limitations in design, were not supportive of an association between shiftwork and breast cancer. There were a relatively limited number of studies (most focused on a single profession, i.e. nurses), some potential for confounding by unknown risk factors, and inconsistent and inaccurate exposure assessments of shiftwork, which may have biased the results towards the null.

Another occupational group of shiftworkers is flight cabin crew personnel, who also experience circadian disruption due to the crossing of time zones. The incidence of breast cancer has been studied in eight cohorts of female flight attendants, all but one consistently reported an increased risk for breast cancer which was greater after a longer duration of employment. Limitations of these studies included the potential for detection bias among female cabin crew due to a higher prevalence of breast cancer screening in this occupational group, proxy measures of exposure used in dose–response relationships, and potential confounding by reproductive factors and cosmic radiation.

The Working Group concluded that the evidence for an association with breast cancer and shiftwork that involves night work was consistent in the studies that were specifically designed to address this question. The studies of cabin crews provided additional support.

Other cancers

Few studies have investigated the association between shiftwork and cancers at other organ sites. Increased risks of cancers of the prostate, colon, and endometrium have been reported. The earliest studies of airline pilots also showed a markedly elevated incidence of prostate cancer compared with national reference levels, but limitations of these studies included the potential for detection bias due to a higher prevelance of screening for prostate cancer in this occupational group.

5.3 Animal carcinogenicity data

Animal models have been used extensively to test the impact of the circadian system (central circadian pacemaker in the suprachiasmatic nuclei and the pineal gland/melatonin-generating system) and its disruption (i.e. phase shifts, light during the dark period, melatonin suppression) on tumour development and growth at all stages of oncogenesis.

SHIFTWORK 761

Two studies examined the impact of continuous high-intensity light versus low-intensity light on tumour development in mice. One study demonstrated clear increases in the incidence of lung adenocarcinomas, leukaemias and lymphomas combined. The second study showed an increase in the incidence of and mortality from mammary tumours in one substrain that had normal vision, and no increase in a substrain of the same strain that had retinal degeneration due to genetic predisposition. A third study showed no effects.

All of the remaining experimental studies used initiation–promotion protocols or tumour growth models following the transplantation of syngeneic tumour fragments, cells, or human cancer xenografts. The species used in these studies included both sexes of rats, mice and hamsters, all of which yielded positive results in at least one study. The types of rodent model systems studied included mammary adenocarcinoma/fibroadenoma, cancers of the peripheral nervous system and kidney, hepatocarcinoma, pancreatic adenocarcinoma, colon adenocarcinoma, prostate adenocarcinoma, squamous-cell carcinoma and fibrosarcoma, osteosarcoma and carcinosarcoma, melanoma, neuroblastoma, and undifferentiated neoplasms.

The model systems used to study the role of the central circadian function and its disruption on cancer development and/or growth encompassed the exposure of animals to chronic alterations in the light–dark environment (i.e. constant bright light, constant darkness, altered light–dark schedules, intermittent light during darkness, single light pulse during darkness). Other model systems used more focused experimental manipulations that included phase-shifting central circadian activity only (i.e. exposure to experimental chronic jet lag), suppression or ablation of the nocturnal circadian melatonin signal (i.e. pinealectomy or exposure to dim light during darkness), ablation of the central circadian activity and of melatonin production (i.e. induction of lesions in suprachiasmatic nuclei), clock gene mutations (i.e. mPer2 knockouts) and the impact of carcinogen administration at different circadian times on tumorigenesis. A specialized model system evaluated the acute proliferative activity of tissue-isolated melatonin-receptor-positive murine or human tumours perfused in situ with different physiological levels of melatonin from natural diurnal blood changes and artificial manipulation.

The major patterns of light–dark environments that have an impact on cancer development and/or growth (i.e. stimulation) are constant light exposure (two positive of three studies, five positive of six initiation–promotion studies, five positive of five tumour-growth studies), dim light during darkness (five positive of five studies), experimental chronic jet lag (two positive of two studies), and circadian timing of carcinogens (four positive of four studies). Two conditions that produced no clear effects or even slowed tumour growth were light pulses during the dark period (two of two studies), and constant darkness (two of two studies). Mechanistically oriented animal studies specifically aimed at investigating the role of the pineal gland (i.e. pinealectomy-induced stimulation of cancer development and/or growth) and the nocturnal melatonin profile (i.e. inhibition of cancer proliferative activity) also had a major impact on cancer (18 positive of 26 studies). Furthermore, a limited number of studies on suprachiasmatic nuclei or clock genes yielded important results with respect to increased tumorigenesis (two positive of three studies).


5.4 Other relevant data

The evidence that relates laboratory investigations and mechanistic considerations to shiftwork-induced carcinogenesis can be divided into two basic fields: disturbance of the circadian system due to light at night with alteration of the sleep–activity pattern leading to potential melatonin suppression and circadian gene alterations; and sleep deprivation that results from the need to sleep when it is not readily possible and misaligned with the surrounding active daytime social environment.

The disturbance of the circadian system is studied at the level of the molecular circadian oscillation. Genes that are responsible for maintaining circadian rhythms have been identified, and may function as transcriptional factors and regulate expression of genes in cancer-related pathways, such as cell cycle, DNA repair, and apoptosis. Animal studies have shown that knockout of the circadian Period gene, Per2, promotes tumour development. Some evidence in humans links genetic polymorphisms in circadian genes to breast cancer and non-Hodgkin lymphoma, and functional loss of the PERIOD genes has been observed in various human tumours. Exposure to artificial light during the night has been demonstrated to disrupt circadian gene expression in mice and humans, which in turn, may alter circadian-regulated biological pathways. Because of their possible roles in tumorigenesis, the light-mediated dysfunction of circadian genes may provide a possible mechanism for the putative carcinogenic effect of light that may or may not involve melatonin.

The light-induced alteration of the circadian system is in part linked to the suppression of melatonin, which is secreted by the pineal gland and acts throughout the organism as a time signal. The suppression of melatonin leads to changes in the gonadotropic axis that specifically involves estrogens and androgens in experimental animals, and may be stimulatory or inhibitory depending on the species and the situation. In humans treated with low pharmacological doses of melatonin, few melatonin-induced changes were documented except a stimulation of prolactin. In animal studies, melatonin in the target sites interferes with the metabolism of estrogen through several metabolic pathways. No data clearly link nocturnal blood levels of endogenous melatonin with endogenous production of estradiol in women.

The decrease in endogenous melatonin may lead to diminished free radical scavenging that may induce local tissue damage, the extent and importance of which is not entirely clear at present.

Melatonin is a direct and indirect immunostimulant; its suppression leads to a state of immunodeficiency that is aggravated by the pronounced effects of sleep deprivation upon the immune system. Prolactin, a strong immunostimulant, is decreased during sleep deprivation.

Direct inhibitory effects of melatonin on tumour cell proliferation have been shown in several animal models not only at pharmacological but also at physiological concentrations.

Sleep deprivation is a common feature in most forms of shiftwork that involves night and/or early morning hours. Changes in the immune system have been shown to occur in

SHIFTWORK 763

partial (early or late night) sleep deprivation and comprise changes in the cytokine pattern that favours the Th2 group of cytokines and decreases Th1 cytokines (e.g. interferon γ) which act in cellular immune defence and in immune surveillance to counteract tumour growth. In the majority of studies of sleep deprivation, suppression of natural-killer-cell activity has been shown, and this also leads to a decrease in anti-tumour surveillance.

The evidence in support of shiftwork-induced carcinogenesis thus links events at the cellular level that affect cell proliferation and endocrine changes with hormonal constellations that promote endocrine-dependent cancers with defects in the immune surveillance that enhance tumour development and growth.

None of these changes stands in isolation; they are all linked to the disruption of the circadian system of shiftworkers and, in combination, may alter the risk of cancer through both tumour induction and promotion.

The experimental data from animal studies in several inter-related physiological systems are strongly suggestive of a causal link between circadian disruption and all its consequences and the development of malignant tumours. Human studies are suggestive of physiological effects that are possibly relevant to carcinogenesis.


6. Evaluation and Rationale

6.1 Cancer in humans

There is limited evidence in humans for the carcinogenicity of shiftwork that involves night work.

6.2 Cancer in experimental animals

There is sufficient evidence in experimental animals for the carcinogenicity of light during the daily dark period (biological night).

6.3 Overall evaluation

Shiftwork that involves circadian disruption is probably carcinogenic to humans (Group 2A).

Date post:	29-May-2020
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

SHIFTWORK...Shiftwork, including night work, involved more than 17% of the total European Union (EU)...

Documents